DP-600 Exam Full Course (6+ hours) | Microsoft Fabric Analytics Engineer

0:00hey everyone and welcome back to the

0:01channel and today we've got a very

0:03exciting video because we're going to be

0:05starting a brand new series here on the

0:08channel we're going to be looking at how

0:10you can become a Microsoft certified

0:13fabric analytics engineer bit of a

0:15mouthful you might know it better as the

0:17dp600 exam that you need to pass if you

0:20want to get this certification now a lot

0:22of people have been asking me about

0:24dp600 how do I get certified what do I

0:27need to learn and so in this course what

0:29I'm going to be doing is bringing

0:31together everything that you need to

0:33know for this certification so that you

0:35can hopefully pass first time so what

0:38are we going to be doing in this video

0:39well this is a bit of an introduction

0:41okay so we're going to be looking at

0:42what's in the exam what are some of the

0:45details of the exam so how long is it

0:47how how can you pass do you do it online

0:49or in person that kind of thing we're

0:51going to look at an overview of this

0:52course so specifically what are we going

0:55to learn in this course in YouTube and

0:57how are you going to learn as well then

0:59we're going to finish by looking at why

1:01you might want to take this course and

1:02why you might want to get fabric

1:04analytics engineering certification in

1:06general so what's in the exam so the

1:09first section is around planning

1:11implementing and managing solutions for

1:14data analytics now within that we've got

1:16various kind of sub modules okay so you

1:20need to understand the requirements how

1:22do we identify requirements for

1:24Solutions how do we do things like

1:26security how do you know the difference

1:28between different data gateways how do

1:30you set up access control and workspaces

1:34and capacities and how do you modify the

1:36settings of all these things as well

1:38another important section of the exam

1:40and being an analytics engineer in

1:42general is Version Control you'll need

1:44to understand how to set up Version

1:47Control for Azure devops understand some

1:50of the settings and the configuration

1:51options that that entails as well as

1:54some of the deployment pipeline

1:56functionality as well now this is not a

1:58completely exhaustive list of what's in

2:00that section of the exam we'll go

2:02through each of those in a bit more

2:03detail it's just the high level kind of

2:05areas that are covered in that section

2:07and this section is worth 10 to 15% of

2:10the exam so it's a small section but

2:12it's a very important section I think

2:14anyway in terms of becoming an analytics

2:16engineer these are really important

2:18topics that you need to understand so up

2:19next we've got 40 to 45% of the exam so

2:23this is really the core of the exam that

2:25you need should probably spend most of

2:27your time studying okay and it's a

2:29around preparation and serving of data

2:33so this is one of the core tasks that

2:35you'll be asked to carry out as an

2:38analytics engineer and within that we've

2:40got quite a lot of really important

2:42topics okay so you've got understanding

2:44The Lakehouse how do we set one up how

2:47do we create tables difference between

2:49tables and files the warehouse so the

2:51tsql experience creating shortcuts

2:54ingesting data from external locations

2:57into our warehouse or our Lakehouse what

3:01are the different methods that we can

3:02choose here and when do we choose

3:04specific ones okay then we've got data

3:06transformation so once we've got our

3:08data within fabric we might want to do

3:10some Transformations on it now you can

3:12do that with tsql spark and you will

3:15need to know at least the basics of tsql

3:17and Spark and also Dax in the next

3:19section as well so there's quite a lot

3:21of languages that you need to know to

3:23kind of an intermediate level I would

3:24say for this exam next we've got

3:26performance so how do we optimize

3:29performance both in terms of the getting

3:32data into fabric but also in terms of

3:34the transformation piece so if you've

3:36got spark jobs that are running really

3:39long or you've got some tsql scripts

3:41that are really not very performing very

3:43well they're taking a long time to kind

3:45of return your data what can you do to

3:47monitor performance then optimize it and

3:49as I mentioned you will need to know P

3:51spark and T SQL to a kind of

3:53intermediate level there so that's

3:55something to bear in mind for this exam

3:57so the next section of the exam is

3:58around implementing and managing

4:00semantic models and this is worth 20 to

4:0325% of the exam and within this category

4:06you've got the different storage modes

4:09so direct query import mode and direct

4:12late mode and kind of understanding how

4:14these things work how to set up direct

4:16Lake mode when to use it maybe when not

4:18to use it you're going to need to have a

4:20good understanding of Dax here there's

4:22quite a few questions around Dax and Dax

4:25studio and tabul editor 2 as well so

4:28these are kind of things that you need

4:29to be be aware of because you'll

4:31probably get some questions around that

4:32as well also within the section you've

4:34got data modeling so things like Star

4:36schemas Bridge tables how do we deal

4:39with many to many relationships and also

4:41we've got things around security how do

4:43we set up roow level security within

4:45your santic model object level security

4:47and how do we validate that that's

4:49actually working correctly so in the

4:50final section of the exam which is worth

4:52around 20 to 25% again is explore and

4:56analyze data so here we're really

4:58looking at the analysis part of being an

5:01analytics engineer because most of your

5:03role might be in around the engineering

5:06piece but really you need to know the

5:08analytics side of thing as well because

5:10that's going to make you a much better

5:11analytics engineer so in this section

5:13you're going to be asked questions about

5:14data analysis specifically using tsql

5:17okay so analyzing your lake house SQL

5:20endpoint analyzing your data warehouse

5:23coming up with insights from data using

5:25tsql you'll also be asked about data

5:27profiling so understanding the pro

5:29profile different tables based on some

5:31of the profiling metrics that you get

5:33and also analyzing data via the xmla

5:35endpoint so that's another part of this

5:38section of the exam so these are the

5:39four sections there's quite a lot to go

5:42through and it covers quite a broad

5:44range of skills that you need to know

5:47from ppar tsql Dax data modeling getting

5:51data in data modeling serving data in

5:53semantic models as well so quite a broad

5:55range endtoend exam so in the exam there

5:58is between 40 and 60 question and the

6:02results are scaled and you get given a

6:04result between 0 and 1,000 so the pass

6:07Mark here is 700 out of 1,000 and that's

6:10the scaled score so that's something to

6:12bear in mind and the questions can be of

6:14different types okay so some of them are

6:16multiple choice some of them might be a

6:18case study and the case studies normally

6:20take two three four questions so you

6:22really need to understand what's going

6:23on and you get multiple questions about

6:25the same case study you also have things

6:27like drag and drop and ordering lists

6:29and for a full list of the question

6:31types recommend you go to this resource

6:33here it's on the Microsoft learn the

6:36exam question section of Microsoft learn

6:39I'll leave a link to that in the

6:40description below you can also use the

6:42exam sandbox that provides you with a

6:44Sandbox experience of the exact question

6:47types that you can expect in the exam

6:49you'll get given 100 minutes to actually

6:51carry out the exam you should set aside

6:53at least 2 hours though so 120 minutes

6:56for the exam just so that you can kind

6:57of get in get settled as I mentioned

7:00it's 700 out of 1,000 to pass this exam

7:03so 70% you can take the exam either in

7:06person in Pon view sensors or you can do

7:09it online so my personal recommendation

7:11would be to take the exam in person if

7:14you can um it kind of eliminates a lot

7:16of the doubts and the problems around

7:19like Wi-Fi and worrying about your desk

7:21setup and your room has to be a very

7:23specific layout and so if you can visit

7:26a center in person I think it's a lot

7:28better cuz you can just walk in take the

7:30exam and walk out whereas online you

7:31have to think about lots of different

7:33things for me it's less stressful to do

7:36it in person if you can so there is an

7:39exam fee which is

7:40$165 for people in the USA it varies for

7:44different countries but if you look on

7:46the right hand side we got a free exam

7:48so if you're very quick and you go to

7:51the link in the description and you

7:53complete the fabric AI skills challenge

7:55training course before the 19th of April

7:58so you've only got a few days you can

8:00get a voucher to take the DP 600 exam

8:04for free so you can do that skills

8:06challenge training very quickly get your

8:09free voucher and then watch all of this

8:11series on YouTube and as long as you

8:13take that exam I think it's before June

8:15the 22nd or something so you've got two

8:17or three months to kind of go through

8:19the material at your own pace and then

8:21you get to save yourself

8:23$165 or the equivalent in your currency

8:26in your country so this is what this

8:28series is going to look like here on

8:30YouTube we've got the first video which

8:31is an introduction to the course then

8:33I'm going to be covering all 11 chapters

8:36of the exam and the content is going to

8:37be delivered through various real world

8:40scenarios CU I want to make this content

8:42interesting for you and also make it

8:44relevant for you if you want to be an

8:47analytics engineer or if you are an

8:48analytics engineer currently in your

8:50career and to do this I'm going to be

8:52combining Theory so I think there is

8:54some theory in some of these modules you

8:56do have to understand how things work

8:58but then practice as well how to

8:59actually implement this stuff in fabric

9:02at the same time throughout the course

9:03I'm going to be reinforcing that

9:05knowledge as we go through and as asking

9:07rhetorical questions as well as sample

9:10questions and at the end I do plan to go

9:12through a full video kind of like a

9:13practice paper let's say designing lots

9:16of questions that you can expect within

9:18the exam as with all of my courses all

9:21of the resources and module notes and

9:23scripts and notebooks and all this kind

9:25of thing I'll be posting that in the

9:27school community so if you're not

9:28already a member there I'll leave a link

9:30in the description below it's completely

9:32for free so make sure you sign up and

9:34yeah you'll get access to all of that so

9:36there's a few reasons that I just want

9:37to cover quickly here around why you

9:39might want to take this course and then

9:41go on to become a certified analytics

9:44engineer well the course gives structure

9:46to your learning Microsoft have given us

9:49a study guide and they've said these are

9:51what we think is important to learn for

9:54an analytics engineer so it gives you a

9:55really good Pathway to follow might also

9:58help you get a new job having that

10:00certification on your CV is definitely

10:02not going to hinder your chances it

10:04would also be good to go into kind of

10:06promotion talks or payiz talks with your

10:08boss and say yeah well last year I did

10:10the certification and this is especially

10:12true if you work in a consultancy right

10:15because consultant here they're kind of

10:16selling your skills and your experience

10:18so if you have certification then that

10:21can help them win work in the future so

10:23in this lesson we've looked at what's in

10:25the exam we've looked at some of the

10:26exam details we've also looked at the

10:29overview of this course so what are the

10:31different chapters that we're going to

10:32be looking through and why I think you

10:34should take this course join us in the

10:37next lesson where we'll be starting the

10:40course properly and we'll be looking at

10:43how to plan and Implement a data

10:45analytics solution hello and welcome to

10:48this first chapter in this dp600 exam

10:52preparation course we're going to be

10:54looking at how to plan a data analytics

10:57environment in fabric now this is the

10:59first chapter in 11 chapters that we're

11:01going to be going through teaching you

11:03everything you need to know to hopefully

11:04pass the dp600 exam in this chapter

11:08we're going to be covering exactly what

11:09you need to know if you look at the

11:11study guide in Microsoft learn these are

11:14the elements that we're going to be

11:15covering how do we identify requirements

11:17for a solution so the various components

11:20features performance capacity skus that

11:23kind of thing how do we make decisions

11:24about that we're also going to be

11:26looking at how to recommend settings in

11:27the fabric admin portal

11:29how do we choose data Gateway types and

11:32also creating custom powerbi report

11:35theme towards the end of the lesson

11:36we'll be testing your knowledge with

11:38five sample questions and just as a

11:41reminder all of the lesson notes and key

11:44points and link to further learning

11:46resources they're going to be published

11:47on the school community so if you're not

11:49already a member I'll leave a link in

11:50the description now you play the main

11:53character in a scenario and this

11:55scenario is going to walk you through

11:57everything you need to know for those

11:58four ele Els of the study guide are you

12:00ready let's begin so you are a

12:02consultant and you're starting your

12:04first day on a new project and this is

12:07Camila she is your client for the

12:10project and on the phone before the

12:11meeting Camila had mentioned that she

12:13wants to implement fabric but she

12:15doesn't know really where to start and

12:17that's where you come in you're going to

12:19start with a requirements Gathering

12:21Workshop so you organize a full day

12:23workshop with Camila the client to truly

12:26understand their business and their

12:27requirements now your goals for this

12:29Workshop are to extract a set of

12:32requirements from the client to help you

12:34build a plan for their new fabric

12:37environment and another goal is to do

12:40such a great job in planning their

12:42environment that the client is going to

12:43give you a new contract by the end of it

12:45to build the thing okay so this

12:47requirements Gathering Workshop what are

12:49you going to ask Camila what do you need

12:52to know when you're identifying the

12:53requirements you should think about

12:55focusing on these three elements to

12:57begin with the capacities so how many do

12:59we need what sizing do the capacities

13:01need to be in this new environment then

13:03we're going to look at data ingestion

13:05methods so there's lots of different

13:06ways that we can ingest data into fabric

13:09you're going to ask a set of questions

13:11that's going to kind of deduce the best

13:13method for getting data into fabric

13:15based on the requirements similarly

13:17we've got data storage so we've got

13:19three different options for storing data

13:21in fabric how do you ask the right

13:23questions and identify the requirements

13:25to choose the right one so let's start

13:27off thinking about capacity requ

13:29requirements now the requirements that

13:30we need here are really the number of

13:32capacities that are required and the

13:34sizing so the SKU the stock keeping

13:37units you probably know by now that in

13:39fabric we have capacities of varying

13:42sizes so what determines the number of

13:44capacities required so from previous

13:46videos you've probably understood that

13:48one of the things that impacts the

13:50number of capacities required is

13:52compliance with data residency

13:53regulations so the capacity dictates

13:56where your data is stored so if you have

13:59regulations that dictate that your data

14:02must reside in the EU for example for

14:04gdpr that's going to be one capacity in

14:06your business if you have other

14:08requirements that say these data sets

14:09need to be stored in the US you're going

14:12to have to have a separate capacity for

14:13that as well another thing that can

14:15impact the number of capacities is the

14:17billing preference so the capacity is

14:20how you get build in fabric so some

14:22organizations might want to separate the

14:24billing between different departments in

14:27their organization so they might have

14:28one capacity for the finance department

14:31one capacity for your Consulting

14:32division one capacity for your marketing

14:34department for example another thing

14:36that could determine the number of

14:37capacities that you need is segregating

14:40by workload type so if you have a lot of

14:42heavy intensive data engineering

14:44workloads then you might want to put

14:46those in a separate capacity and give it

14:48enough resource to allow you to do that

14:51in a confined capacity then you're

14:53serving of business intelligence you

14:55might want to do that in a separate

14:56capacity so that the read performance on

14:58those kind of dashboards is not impacted

15:00by the heavy data engineering stuff

15:02maybe machine learning stuff that's

15:04being done in other capacities you might

15:06also want to segregate by department

15:08just through business preference as well

15:09aligned with that billing preference so

15:11some companies like to have their

15:13capacity aligned to various dep

15:14departments within their business so

15:16these are the things you need to extract

15:18in terms of requirements when you're

15:19talking with this client and what about

15:21the sizing well we've touched on that

15:23already but some of the things that

15:24impacts the sizing of a capacity are the

15:27intensity of the expected workloads so

15:30are you going to be doing High volumes

15:31of data ingestion are you going to be

15:33getting gigabytes of fresh data into

15:36Fabric or even terabytes of data into

15:38fabric every day these are going to use

15:40a lot of your resources and to go

15:42through them quickly it helps if you

15:44have a higher capacity similarly heavy

15:46data transformation so if you're doing a

15:48lot of heavy transformations in spark

15:51that's going to use a lot of resources

15:52so if that's something you're going to

15:53be doing regularly in your business you

15:56want to be choosing a high capacity for

15:57that again machine learning training can

16:00you be very resource intensive going to

16:02take hours or sometimes even days to

16:04train a machine learning model if that's

16:06something you're going to doing

16:07regularly you want to be having that on

16:08a high capacity the budget of your

16:11client also dictates the capacity the

16:14sizing of the capacity that you're going

16:16to choose obviously the more resources

16:18the higher that SKU that you decide the

16:21more expensive it's going to be and some

16:23clients might be very sensitive around

16:25the cost and related to that is can you

16:27afford to wait or can the client afford

16:29to wait because if you procure a F2 skew

16:33it's probably going to go through your

16:34data but it might take a very long time

16:37and in some business that might not be a

16:38problem maybe you're just doing data

16:39ingestion once per day you ingest all of

16:42your fresh data and it might take a lot

16:45longer on an F2 capacity but that's not

16:47necessarily a problem maybe you can do

16:48it overnight and by the time people come

16:50in in the morning all of that data has

16:52been ingested or transformed and it's

16:54ready for consumption in the morning so

16:56what's your propensity to wait now some

16:59other companies might have regular data

17:01coming in every hour like gigabytes of

17:04data every hour and in that scenario you

17:06really need a high capacity to be able

17:08to churn through all of that stuff and

17:10get it processed before the next hourly

17:14load for example another thing that can

17:16determine the sizing of the capacity is

17:19does the client want access to f64

17:22Features so there's quite a lot actually

17:24of features that open up when you get to

17:27f64 so co-pilot being a good example

17:30currently and there's many many more

17:32I'll list them on the screen here these

17:34are features that only really are

17:35available if you choose f64 capacity or

17:39above so that's something to bear in

17:40mind if you want to use any of these

17:42features you need an f64 plus so what

17:44about the data ingestion requirements

17:46well here what we really need to know is

17:49what are the fabric items and or

17:51features that you need to get data into

17:53Fabric and how are you going to

17:54configure these items once you've built

17:56them now some of the options here and

17:58this is not an exhaustive list there's

18:00lots of different options here we've got

18:02the shortcut database mirroring ETL via

18:05data flow ETL via data Pipeline and a

18:07notebook and the event stream so these

18:09are some of the options that you might

18:11want to consider so what are the

18:13questions that you need to ask of a

18:15client when you're identifying the

18:17requirements to help you make the

18:19decision here well these are some of the

18:20deciding factors the main one really is

18:22where is the external data stored if

18:25it's in ADLs Gen 2 Amazon S3 or S3

18:29compatible storage location like Cloud

18:31flare for example Google Cloud Storage

18:33or the data verse but then these are the

18:35ones that are going to be available for

18:37you to shortcut into fabric so if you

18:40get any questions in the exam around you

18:42know my data rest stored in ADLs Gen 2

18:45well obviously the shortcut is a good

18:47option for that now it's not necessarily

18:48the only option you can still do ETL via

18:52any of these storage locations but it

18:53does open up that shortcut possibility

18:56now if you see Azure SQL Azure Cosmos DB

18:59or snowflake mentioned then immediately

19:01you should start thinking okay this

19:02could be database mirrored so you can

19:05use database mirroring to create that

19:07kind of live link to the database and

19:09it's going to maintain a mirror inside a

19:12fabric is it on premises now if you're

19:14data stored on premises then you're

19:17going to be probably want to be using

19:18the ETL via data flows or data pipelines

19:22because these two activities these two

19:24items allow you to create that

19:26on-premise data Gateway on your on

19:29premise server and then connect to that

19:30via the data flow or the data Pipeline

19:32and if you got realtime events realtime

19:34streaming data obviously you probably

19:36want to using the event stream to get

19:38that data into fabric anything else

19:40really you're going to be looking at ETL

19:42by either the data flow the data

19:44Pipeline and the notebook and when to

19:46choose which one well I've done a very

19:48long video I'll leave a link in the

19:49description or you can click here to

19:51make that decision about which of these

19:53is best for that particular organization

19:56so related to that is also what skills

19:58exist in the team because you don't want

20:00to build a solution that can't be

20:02maintained managed by the company or

20:04your company or your client's company so

20:06if you're looking for a predominantly no

20:08and low code experience then you're

20:10going to want to be focusing on the ETL

20:13via data flows and data pipelines both

20:15of these are fairly low and no code

20:18experiences help you get data into

20:19fabric if You' got a lot of SQL

20:21experience in your team then here you

20:23can be using the data pipeline you can

20:25use the script activities to do

20:27Transformations on your data as is

20:28coming in and if you have people that

20:30are familiar with spark python Scala

20:33that kind of thing then you can use the

20:36ETL notebook if you're you know perhaps

20:39you've got data coming from a rest API

20:41and you want to be using python

20:42libraries to get that in that's a good

20:44option for you there so whil we're on

20:46the topic of data ingestion there's a

20:48few other features you need to be aware

20:50of that might come up in the exam that

20:53can help you identify different

20:54requirements for getting data into

20:55fabric these are the on premise daily

20:58Gateway which we've mentioned the v-net

21:00the virtual network data Gateway fast

21:02copy and staging so you might be asking

21:05some questions about these things in the

21:06exam as well so when do we decide on

21:09these sorts of things well you need to

21:11ask how the data in the external system

21:13is being secured right so if it's on

21:16premise if it's an on premise SQL Server

21:18you have to be using the on premise data

21:20Gateway if your data is living in Azure

21:23behind some sort of virtual Network or

21:25private endpoint that kind of thing then

21:28you want to be setting up the v-net data

21:30gateway to access that and in terms of

21:31the volume of data this is also going to

21:33have an impact on the items that you

21:36choose for doing your data ingestion and

21:38also some of the features available so

21:40if you've got low or medium data per day

21:43well if it's low then you probably don't

21:44need any of these specific features like

21:46the out of the box Solutions will be

21:47good enough but if you've got quite a

21:49lot of data gigabytes per day in that

21:52kind of range you want to be using some

21:54of the features like Fast copy and

21:56staging similarly if You' got very high

21:58amounts of data these are going to be

22:00one of using the fast copy and the

22:01staging if you're using data flows

22:03alternatively you can use data pipelines

22:06and if you can get data in bya a fabric

22:08notebook then that's another option as

22:10well so before we move on I just want to

22:12mention a bit more detail around the

22:14data gateways now as you probably know

22:17already there are two types of data

22:19gateways that we can configure in

22:21Microsoft fabric number one is the

22:23on-premise data Gateway and number two

22:25is the virtual network data Gateway and

22:28a data Gateway in essence helps us

22:30access data that's otherwise secured so

22:34if his data is on an on-premise SQL

22:36server for example it gives us a secure

22:38way to access that data and bring it

22:40into fabric likewise if you've got data

22:43behind a virtual Network secured in

22:45Azure in like blob storage or ADLs Gen 2

22:49it provides us with a secure mechanism

22:51to access that data so I'm not going to

22:53show you step by- step how to set up a

22:55data Gateway in this lesson but what it

22:58done is linked to two other videos by

23:00other creators that show the process in

23:02detail if you want to go and have a look

23:04I'll leave that in the school Community

23:05but I do want to just cover kind of the

23:07high level process for each of them just

23:09so you understand a bit more about what

23:11that looks like if you've never set one

23:13up before so for the on- premise data

23:15Gateway there's a few high level steps

23:18number one we need to install the data

23:19Gateway on the on premis server and if

23:22you've already got an on- premise data

23:23Gateway set up on your on premise server

23:26perhaps you're using it in traditional

23:28powerbi data flows for example then

23:30you're going to need to update it to the

23:32latest version cuz that's going to be

23:33compatible with Microsoft fabric the

23:36next step is to in fabric create a new

23:39on premise data Gateway connection and

23:41then from that you can connect to that

23:43data Gateway from either a data flow and

23:46now also a data pipeline so the data

23:48pipeline was recently added in the last

23:49few weeks I think it's still in preview

23:52that connection so you might not get

23:54asked about it in the exam but it's good

23:55to know that now it's actually possible

23:57via the data flow and the data pipeline

23:59to set up the v-net data Gateway we're

24:01going to start in Azure there's a few

24:03settings that you need to configure in

24:05your Azure environment before you can

24:07set up the v-net data Gateway connection

24:10so you're going to need to register a

24:12Power Platform resource provider within

24:14your Azure subscription and then within

24:16the item that you want to share or you

24:18want to access for example in your Azure

24:21blob storage item in Azure you need to

24:23create a private endpoint in the

24:24networking settings then create a subnet

24:27and then we're going to use that in

24:28fabric to create a new virtual network

24:31data Gateway connection and then again

24:33from that you can connect to it via your

24:34data flow to be able to access that data

24:37that is behind that virtual Network in

24:40Azure so next let's look at the data

24:42storage requirements and when we're

24:44talking to our client here identifying

24:46requirements really what we're trying to

24:48extract is okay what fabric data stores

24:52are going to be best for these

24:54requirements and what overall

24:56architectural pattern are we going to be

24:58aiming for with this solution now the

25:00options here are obviously The Lakehouse

25:02the data warehouse and the kql database

25:05and some of the deciding factors to

25:07choose between these are well what's the

25:09data type okay so is it structured or

25:12semi structured or even unstructured so

25:15are you going to be getting raw files

25:18CSV Json maybe from AR rest API is it

25:21unstructured is it image data video is

25:24it audio data for example these are all

25:27going to be wanted to store in The

25:28Lakehouse because this is kind of the

25:30only place in fabric where you can store

25:32a variety of different file formats if

25:35your data is relational and structured

25:37then obviously you can keep that in

25:39either the lake house or the data

25:41warehouse and if it's real time and

25:42streaming you're going to be want to

25:43streaming that into your kql database

25:46next up another important consideration

25:48when choosing a data store is what

25:51skills exist in the team so if you're

25:53predominantly tsql based then you're

25:55going to be want to be using the data

25:57warehouse experience

25:59if you're predominantly spark and python

26:01Scara that kind of thing then you're

26:03going to be wanting to storing your data

26:04predominantly in the lakeh house and if

26:06you're predominantly using kql in your

26:08organization that's going to be one to

26:10using kql database for your data storage

26:14congratulations you've completed your

26:16first engagement for Camila you've

26:18convinced her to set up a proof of

26:20concept project in her organization so

26:23she's already created the fabric free

26:25trial she's set up her environment but

26:28immediately she's hit a bit of a hurdle

26:30so this is your next mission she Rings

26:32you up and she says hey I need some help

26:34I open the fabric admin portal and

26:37nearly had a heart attack please can you

26:39help me understand all of these settings

26:41so you set up a call with Camila to help

26:43her understand the fabric admin portal

26:46how are you going to teach her and what

26:47are you going to teach her about the

26:49admin portal what are the most important

26:51settings that she needs to know about

26:52okay just before we get into the fabric

26:54admin portal and look at some of the

26:56settings available to us in there there

26:58it's important to note that to be able

27:00to access the admin portal of course

27:02first you need a fabric license but then

27:04you need to have one of the following

27:07roles you need to be either a global

27:09administrator a Power Platform

27:11administrator or a fabric administrator

27:14So within the admin portal here in

27:17fabric you'll see this menu on the left

27:20hand side so these are some of the

27:21important settings in tenant settings

27:24here you can allow users to create

27:25fabric items so if you just set up

27:27fabric in your ization you need to allow

27:29people to actually create fabric items

27:31without that you can't really get very

27:32far enable preview features so every

27:35time Microsoft release new features

27:37normally they put them in the admin

27:39portal and you can allow or disallow

27:42users in your organization to use them

27:44you can also allow users to create

27:46workspaces there's a whole host of

27:48security related features that you can

27:51manage and get control over in your

27:53tenant so for example how do you manage

27:55guest users allowing single sign on

27:57options for things like snowflake big

27:59query red shift accounts that kind of

28:01thing how do you block public internet

28:03access so that's really important to

28:05know enabling other features like Azure

28:08private link for example allowing the

28:10service principal access to the fabric

28:12apis so if you're going to be doing some

28:13automation you need to allow access to

28:17service principles to the API there's

28:19also options in there for allowing git

28:21integration so if you're setting up

28:22Version Control that needs to be enabled

28:25there and there's also some features

28:27like allowing cop pilot within the

28:28organization as well now in general some

28:31of the settings can be one of three

28:33things it could be enabled for the

28:35entire organization it can be enabled

28:37for specific security groups so say you

28:39only want super users to be able to use

28:43this feature or admins within your

28:45fabric environment to use a specific

28:47feature then you can enable it for

28:48specific security groups or you can

28:50enable it for all except certain

28:52security groups so everyone in your

28:54organization gets access apart from

28:57these people perhaps guest users is a

29:00good example now other settings in the

29:02fabric tenant settings are kind of

29:04binary you either enable them or you

29:06disable them for the entire organization

29:08another important point in the fabric

29:10admin portal are the capacity settings

29:13so this section here and in here you can

29:16create new capacities delete capacities

29:18manage the capacity commissions and also

29:21change the size of a capacity so these

29:23are some important capacity settings

29:25that you need to be aware of understand

29:27how they work and how to manage them

29:29within your fabric environment so great

29:31you've talk Camila about the fabric

29:33admin portal and she's very grateful but

29:36before the meeting ends she has one more

29:38thing she wants to ask you about she

29:40says one final thing before you go and

29:42it might seem a bit random but when we

29:44migrate to fabric I want our bi team to

29:47create more consistent reports have you

29:50got any ideas about how we can achieve

29:51that and of course the first thing you

29:53think of are custom powerbi report

29:55themes now there are many ways to create

29:58a custom report theme in powerbi you can

30:01either update the current theme if

30:03you're in powerbi desktop or you can

30:05write a kind of Json template yourself

30:07using the documentation if you're

30:08feeling a bit Brave you can do that

30:10yourself or you can also use a third

30:12party online tool there's quite a few

30:14report theme generator tools that exist

30:16online but it's unlikely you're going to

30:18be tested on that in the exam so your

30:20task is to show Cilla how to create a

30:22custom report theme so let's have a look

30:25at how you can do that within powerbi

30:27desktop top so here I've got a report

30:30and what I'm going to do to access the

30:31report themes you need to go to the view

30:33tab then you can see these themes here

30:35and obviously these are the preset

30:37themes so you can just click and update

30:39the current theme very simply like that

30:41but to do most of the customization you

30:43need to click on this button here and

30:45you can access the current theme and all

30:47accessible themes are currently

30:49installed on this machine then there's a

30:51few settings down here that quite

30:52important to know so browse for themes

30:55if you click on that it's going to allow

30:56you to import a powerbi report theme so

31:00if you've already got a theme Here For

31:01example this one here then you can

31:03select that and install it into your

31:05environment like so if you want to

31:07customize the current theme you can do

31:10that like this and it's going to bring

31:11you through to this UI environment just

31:13to you know change some colors change

31:15some text change some visuals what you

31:18have to think about for this section of

31:20the exam is what could they ask you you

31:22have to think about how could you

31:23possibly be tested on this so in terms

31:26of the powerbi report theme stuff you're

31:28likely to be tested on these buttons

31:30here and what they do plus they could

31:32ask you about a Json theme so they could

31:35show you a Json theme and maybe ask you

31:38about okay how can you edit this theme

31:40what doesn't look right in this theme

31:42that kind of thing so it's good to have

31:44a bit of familiarity about the different

31:47sections in these Json files so the name

31:50of it how you can store your data colors

31:52as a list some of these different

31:54settings here you're probably not

31:55expected to memorize all of the

31:57different settings in Json format but

31:59you might get shown a theme in Json

32:01format and asked to modify it or asked

32:04to comment on it in some way to export

32:06the current theme you can also use this

32:08save current theme and that's going to

32:09allow you to export a Json file that you

32:11can share within your organization and

32:13you've also got here access to the theme

32:15Gallery so this is going to bring you

32:17through to the theme Gallery website

32:19where you can download other people's

32:21themes for your report to finish up this

32:23video and this lesson we're going to go

32:26through five practice questions just to

32:28kind of solidify that knowledge make

32:29sure you're understanding some of the

32:31key Concepts within a context of a

32:33scenario so the first question is you're

32:36running an F2 capacity and you regularly

32:39experience throttling with that capacity

32:41now there's a number of long running

32:43spark jobs that take on average 3 hours

32:45to complete and you need these to

32:47complete in under 1 hour so you plan to

32:49increase the SKU of the capacity where

32:52would you go to make this change would

32:53you go to the workspace settings and

32:55configure spark settings would you go to

32:57to the admin portal and the capacity

32:59settings section and then click through

33:01to Azure to update your capacity would

33:03you go to the monitoring Hub and look at

33:05the Run history or would you use the

33:07capacity metrics app so pause the video

33:09here and have a bit of a think and then

33:12I'll move on so the answer is 2b so you

33:15can manage your capacity settings within

33:16admin portal and then capacity settings

33:19and then you can actually click through

33:21to Azure it gives you a link to the

33:22Azure portal and that's where you're

33:24going to change the capacity within the

33:26Azure portal OB so you can't be within

33:28the spark settings that's for managing

33:30the configuration of your spark cluster

33:32within a workspace and in the monitoring

33:34Hub we can't get anything there to do

33:36with capacity settings that's just going

33:38to tell you how your jobs are running

33:39and in the capacity metrics app that's

33:41just a readon app for having a look at

33:44how your capacity is being used so that

33:46wouldn't also be suitable either

33:48question number two your data governance

33:50team would like to certify a semantic

33:52model to make it discoverable in your

33:55organization now only the data

33:56governance team should be able to do

33:58this in what order should you complete

34:00the following tasks to certify a

34:02semantic model so have a look at the

34:04five actions here and what you're going

34:07to have to do is put these in an order

34:10so these are this is an ordered list it

34:12should be so one of the things you'll

34:13have to do first second third fourth and

34:16fifth so once you've got these in order

34:18we'll move on so let's look at the

34:19answer now so the correct order looks a

34:21bit like this so we start by creating a

34:25security group for the data governance

34:27team the clue in the question was that

34:30only the data governance team should be

34:32able to do this so when you see that you

34:35think okay well they need to be within a

34:37security group to enable this number two

34:40and you could argue that one and two

34:41could be interchangeable but these are

34:43the first two items anyway but enable

34:45the make certified content discoverable

34:48So within the admin portal the tenant

34:51settings as a section for Discovery

34:54you're going to need to enable that for

34:55the organization and then after that

34:57going to have to make sure that that

34:59settings is applied only to the data

35:01governance security group that you set

35:03up then you're going to need to ask the

35:05data governance team to go into the

35:08semantic model settings and then

35:09endorsement and Discovery and click

35:12certify for that semantic model and then

35:14you want to validate that that has been

35:16set up correctly and your business user

35:18can see that certified semantic model

35:22within the one leg data Hub three you

35:24join a new company and you're given a

35:26powerbi report theme as a Json file to

35:28use for all new projects how do you

35:30apply this Json file theme to the report

35:33that you're currently developing is it a

35:35in powerb desktop go to view themes and

35:38customize current theme B go to the

35:40fabric admin portal click on custom

35:43branding and then set the default report

35:45theme C use tabulate editor 2 to update

35:48the theme or D in power by desktop go to

35:51the view themes and then browse for

35:53themes so the answer here is D in power

35:57desktop go to the view themes and then

35:59browse for themes so d and a are quite

36:02similar but a is for customizing a

36:05current theme so that's not going to be

36:06allowing you to import a Json file

36:09that's going to allow you to use the

36:11user interface to update the current

36:13theme so that's not what we want to do

36:15we want to import ad Json file as our

36:17report theme which is possible using D B

36:20that functionality doesn't actually

36:21exist custom branding does exist but

36:23that allows you just to update the

36:25colors and the icons within fabric not a

36:28default report theme and tabul editor 2

36:31is also the incorrect answer question

36:33four you have 1,000 Json files stored in

36:36Azure data Lake storage ADLs Gen 2 that

36:39you want to bring into fabric the ADLs

36:42Gen 2 storage account is secured using a

36:45virtual Network which of these actions

36:48would you need to perform first is it a

36:51in fabric go to manage connections and

36:53gateways and then click on create a new

36:56virtual network data Gateway B create a

36:58shortcut to the ADLs Gen 2 storage

37:01account C in Azure register a new

37:03resource provider and create a private

37:05endpoint and subnet or D install an on

37:08premise data Gateway on an Azure virtual

37:10machine in the same virtual Network or E

37:13enable Public Access in the storage

37:15account network settings so for this one

37:17you'll remember that the answer is C so

37:20the first step in setting up a virtual

37:22network data gateways well we need to go

37:24into Azure we need to perform some

37:26network configuration okay so you need

37:28to register that new resource provider

37:31that Microsoft Power Platform resource

37:33provider within your subscription and

37:35then on the item create a private

37:36endpoint and a subnet all of the other

37:38options some of them are steps in the

37:41process but not the first step so the

37:43question was which of these actions

37:45would you need to perform first so yes

37:48we do need to do a but it's not going to

37:50be the first thing that you're going to

37:51do B is kind of a bit of a a red herring

37:54here cuz you might have seen ads Gen 2

37:56and thought ah shortcut but actually you

37:58need to configure the virtual Network

38:00daily Gateway before you can even think

38:03about kind of connecting to it D

38:04installing the on- premise data Gateway

38:07well you know that we're looking at a

38:08virtual Network here so you're going to

38:10be choosing the virtual network data

38:12Gateway rather than an on- premise Ste

38:13goway and E enable Public Access in the

38:16storage account network settings while

38:18that's going to expose your data to the

38:20public internet so not advisable

38:23question five you have data stored in

38:25tables in Snowflake which of the

38:27following cannot be used to bring the

38:29data into fabric a use the data pipeline

38:32copy data activity B create a shortcut

38:34to the snowflake tables from your Lake

38:37housee C use the data flow Gen 2 with

38:40the snowflake connector D use database

38:42mirroring to create a mirrored snowflake

38:45database in fabric so the answer here is

38:47B to create a shortcut to the snowflake

38:50tables from your lake house as you'll

38:52know you can only shortcut to ADLs Gen 2

38:55or Amazon S3 or Google Cloud Storage so

38:59the ability to shortcut is generally on

39:02files when we're talking about tables in

39:05databases whether there all of the other

39:07three we can use so you can do a copy

39:09data activity from a data pipeline to

39:11bring that data in if you want to copy

39:13it in or you can use a data flow Gen 2

39:16or you can use database mirroring

39:18because snowflake is one of the

39:20databases where database mirroring is

39:22possible Camila says thanks she's

39:24seriously impressed with your knowledge

39:26well done in this lesson we've looked at

39:28how you can identify requirements for a

39:30fabric solution we've looked at the

39:32different types of data gateways that

39:34are available to us in fabric we've

39:36looked at the settings in the admin

39:38portal and we've also looked at how to

39:40create custom powerbi report themes and

39:43the good news is you want an extension

39:44to the contract Camila would like you to

39:47implement and manage her data analytics

39:49environment so you've got the next stage

39:51of the contract in the next lesson we'll

39:53look at how you can do that how you can

39:55set up Access Control sensitivity

39:58labeling workspaces capacities all that

40:01kind of stuff how do we set these things

40:03up inside fabric so make sure you click

40:05here for the next lesson hey everyone

40:07welcome back to the channel today we're

40:09going to be continuing our dp600 series

40:12and we're going to be looking at

40:13implementing and managing a data

40:16analytics environment this is the second

40:18part of the dp600 syllabus or the study

40:21guide that we're going through on the

40:23channel and today we're going to be

40:24going through these particular elements

40:26and these are coming straight from the

40:28study guide so number one we're going to

40:30be looking at implementing workspace and

40:32item level Access Control implementing

40:34data sharing for workspaces warehouses

40:37and lake houses managing sensitivity

40:39labels configuring fabric enabled

40:42workspace settings and managing fabric

40:45capacity we've got five sample questions

40:47so at the end of the video we'll be

40:48going through some sample questions to

40:50test your knowledge and as with the last

40:52video I'll be posting the key points and

40:55links to further resources in the school

40:57Community available for free I'll leave

40:59a link in the description for that we'll

41:01continue the scenario that we were

41:03developing in the last lesson you're the

41:05main character again are you ready let's

41:07begin so just to recap you are a

41:09consultant and you're working with your

41:10client who is called Camila and in the

41:12last engagement you successfully planned

41:15their data analytics environment now

41:17you've won the contract to support

41:18Camila in implementing that solution so

41:21Camila is busy doing her resource

41:24planning she's thinking about who she's

41:26going to need to support support this

41:27environment in fabric she's asking for

41:30your assistance to help her structure

41:32her thoughts and also her team so let's

41:34have a look at what that looks like in

41:36fabric so this is a high level structure

41:39of a fabric implementation now you

41:42notice that it's hierarchical at the top

41:45we have tenant level so this is kind of

41:47the one tenant that you're going to have

41:49in your organization then below that you

41:51might have one or many capacities and we

41:54talked about capacities in the previous

41:56lesson we're going to going to be doing

41:57a bit more on capacities in this lesson

41:59as well then in each capacity you might

42:01have one or multiple workspaces then in

42:04each workspace we go down to the item

42:07level so you might have a data warehouse

42:09and a Lakehouse in your workspace then

42:11we can actually go one level deeper than

42:13that which is called the object level So

42:15within the data warehouse you have dbo

42:17do customer that might be a table in

42:19your data warehouse or a view in your

42:21data warehouse that's at the object

42:23level now when we're administering

42:25fabric we need to be aware of these

42:27different levels because at each of

42:28these different levels Administration

42:31happens in a different way in the last

42:33lesson we looked at the tenant level

42:35admin settings so that's mostly things

42:37in the admin portal under the tenant

42:40settings section today we will explore

42:42item level a little bit later on but

42:45first I just want to look at how we can

42:47administer each of these three top

42:49levels so here we have a table now the

42:52table isn't complete yet we're going to

42:54walk through it together so we have the

42:55three top levels we got tenant level the

42:58capacity level and the workspace level

42:59and then on the right hand side we're

43:01going to go through what the

43:02administrator or who the administrator

43:04is what role they require and also where

43:07the admin happens right so where are

43:09they going to be working where are the

43:11settings that they need to administer at

43:13each level so starting at the top level

43:15we've got the tenant admin now to get

43:18the rights to be able to be a tenant

43:20admin we're actually going to go higher

43:21than fabric we need an entra ID role of

43:24global administrator Power Platform

43:26administrator or fabric administrator

43:28and we looked at what that looks like in

43:30the last lesson and they're going to be

43:32working predominantly in the fabric

43:34admin portal so if you have any of these

43:36three roles that's going to be available

43:38to you and you can configure your tenant

43:40settings in there one level down at the

43:43capacity level well the capacity admin

43:45is assigned when you create a new fabric

43:48capacity in Azure now where the

43:50administration happens so any sort of

43:53administration at that capacity level is

43:55going to be done either in the Azure

43:57portal as we mentioned before or in the

44:00fabric admin portal there's a section on

44:02capacity settings we're going to have a

44:04look at both in a minute at the

44:05workspace level so we're going one level

44:08down now you're going to have a

44:09workspace admin they're going to be the

44:11person that is kind of in charge of that

44:13admin level and the role required here

44:16is a workspace role so it's going to be

44:18a person or a group with the workspace

44:21role of admin and again we'll have a

44:23look at what that means in more detail a

44:26bit later on now there going to be doing

44:27most of their Administration within the

44:30workspace settings and also within the

44:32manage access so these are the two areas

44:34that they're going to be focusing most

44:36of their time on now we looked at the

44:38tenant level admin settings previously

44:40in the last video in this lesson we're

44:41going to focus on the capacity level

44:44settings and then the workspace level

44:46settings so let's just focus in on the

44:48capacity administrator settings for a

44:50moment and as I mentioned there's two

44:52really places where capacity

44:55Administration gets done number one one

44:57is in Azure because we need to use Azure

45:00portal to purchase capacity right so

45:04that's going to be where you go to

45:05create a new capacity delete existing

45:07capacities changing the size of a

45:10capacity so if you've got an F2 skew and

45:13you want to go up to an F4 that's going

45:15to be done in the admin portal and also

45:17changing the capacity administrator so

45:18you can do that within the Azure portal

45:20as well as well as that within fabric we

45:24can change some of the settings for a

45:26particular Capac

45:27so we can do things like enabling

45:29Disaster Recovery viewing the capacity

45:31usage report so how much is our capacity

45:34being used we can Define who can create

45:36workspaces within that capacity we can

45:38Define who is a capacity administrator

45:41we can update the powerbi connection

45:43settings so who can connect to this

45:46capacity or items within this capacity

45:49from powerbi and how does that look like

45:51we can permit workspace admins to size

45:54their own custom spark palls so this is

45:57quite important right because you might

45:58want to set some sort of limits on the

46:00sizing of the spark poles that workspace

46:03owners underneath the capacity or in

46:06this capacity you might want to limit

46:08how high they can go with their spark

46:09poles because that's going to have quite

46:10a big impact on the overall capacity

46:13usage so if you're sitting at the

46:14capacity level you might want to add

46:16some restrictions on you know the custom

46:18spark configurations that happen in Your

46:21Capacity and we can assign workspaces to

46:24the capacity in this section as well

46:26okay so so here we are in the portal and

46:28I just wanted to show you some of the

46:30capacity settings how to administer a

46:33fabric capacity and we're going to start

46:34right from the beginning so how do you

46:36actually set up and buy a capacity well

46:38you go to Microsoft fabric if it's not

46:40already there then you can just search

46:42for it here Microsoft fabric this is

46:43going to bring you through to the the

46:45resource creation tool we're going to

46:47click on Create and we're going to walk

46:49through these steps to create a fabric

46:50capacity so you need a subscription and

46:53a resource Group and you can enter the

46:55capacity name so give it a region change

46:58the size I'm just going to do an F2

47:01select and here is where you sign the

47:02capacity administrator and again we can

47:04change that afterwards but that's you

47:06need at least one to set it up so it's

47:08going to give you the estimated cost per

47:10month here and then you press create and

47:12it's going to deploy that capacity okay

47:15so now my capacity has been created we

47:18can click on go to Resource and this is

47:20where we're going to do some of the

47:21administration tasks within the Azure

47:24portal right so here we can have a look

47:26at well firstly we can pause it so if

47:28you want to pause the capacity you can

47:30do that here delete it as well down the

47:32left hand side we've got some useful

47:34things here so capacity administrators

47:36that's where you're going to change your

47:38capacity administrator we've also got

47:40change size so if you're finding that F2

47:44is not enough for your workloads that

47:47you're running you can change it to F4

47:49or f256 if you've got a spare 40 Grand a

47:52month to be using on fabric I'm going to

47:55keep it as an F2 that's just something

47:57to bear in mind there okay now I just

47:58want to have a look at what that

47:59capacity setting looks like inside

48:02fabric so if we go to the admin portal

48:05and capacity settings if we go over to

48:07the fabric capacity tab here we can see

48:10that fabric capacity that we've just set

48:11up so it's an F2 it's in UK South and

48:14it's active so we've got some actions

48:16here we can change the name we can have

48:17a look at the admin you can't really do

48:19much there there's a link through back

48:21into Azure so if you want to make any

48:23changes to it from within fabric you

48:25need to click on this link here but if

48:26you click on the actual capacity name we

48:28go through to the capacity settings

48:31right so this is where you're going to

48:32be doing things like enabling Disaster

48:35Recovery having a look at the usage

48:37report for that capacity turning on

48:39notifications so it's going to give you

48:41a notification when you've used x amount

48:44of percent of your capacity updating who

48:46is the administrator for that capacity

48:49can also be done here changing how we

48:51can access powerbi and how powerbi can

48:54access data in the capacity we've also

48:57got data engineering settings it's

48:59mainly spark settings and this is where

49:02we're going to have that permission to

49:04permit people to change the custom spark

49:07poing right so either on or off and we

49:10can assign certain workspaces to this

49:12capacity so if you just create a new

49:13capacity it's going to be empty and you

49:15can move existing workspaces onto that

49:18capacity so as a workspace administrator

49:20we've got a few different options so

49:22here we're stepping down level into the

49:24workspace level and a workspace

49:26administrator as we mentioned deals

49:28primarily in the workspace settings and

49:31here you can edit the license for the

49:33workspace so change it from like for

49:36example a trial capacity to a fabric

49:38capacity or from Pro to PPU premium per

49:41user we can also configure connections

49:43to Azure as well as well as configuring

49:46Azure devops connections so if you want

49:48to use Git Version Control for this

49:51workspace that's going to be done in the

49:53workspace settings another thing we can

49:54do in the workspace settings is set up

49:57what's called a workspace identity now

49:59this is basically having a managed

50:02service principle dedicated for your

50:04workspace and it basically means you can

50:06connect to things like ADLs Gen 2 for

50:09things like shortcuts and you can do

50:11that in a kind of trusted workspace

50:13access manner basically this is another

50:15pretty new security feature that they've

50:17added quite recently and I'll be going

50:18through more of the security principles

50:20in more detail probably in a separate

50:22video at the end of this series CU

50:24they're quite important we can also edit

50:26some of the power settings in the

50:27workspace settings and also the spark

50:30settings so particular default

50:32environments that we might want to set

50:34up within this workspace things like

50:36that just note here that managing access

50:38is done through the the managing access

50:41section so it's slightly set it's in the

50:43same kind of area but it's not

50:44necessarily in workspace settings where

50:46we add users and add groups into our

50:49workspace we'll have a look at that in a

50:51bit more detail okay next I just wanted

50:52to walk through some of the workspace

50:55settings in a bit more detail so here we

50:57are in a workspace it's called Share Hub

50:59what we're going to be doing is clicking

51:01on this Dot and you can see that we've

51:03got two here that are useful for

51:05workspace administrators number one is

51:07managing access so this is how we're

51:08going to give people access to our

51:10workspace either person or a group we

51:13can add people in here and we can give

51:15them admin member contributor or viewer

51:17if we click on these dots again and then

51:19go through to the workspace settings

51:21this is where we're going to be able to

51:23edit some of the settings for our

51:24workspace General is you just do the

51:27image and the description and also

51:28domains if you're using domains license

51:30info so this is where you're going to

51:32change the potential capacity and the

51:34license that's being used in that

51:36workspace so this one is a trial

51:37workspace so maybe we want to actually

51:39change that to an F2 maybe we've got

51:41some fabric capacity we can select which

51:44one we want to use I'm going to be using

51:46this fabric F2 learn capacity and that's

51:48going to change the license for the

51:50workspace we've also got connecting to

51:53Azure connecting to git downloading

51:55things like the file explor

51:57as well and enabling caching for

51:59shortcuts is another workspace setting

52:01we've got here managed identities so if

52:03you're on an f64 capacity or higher you

52:06can make use of workspace identities and

52:09that's going to basically allow you to

52:10create kind of like a managed service

52:12principle just for this workspace so

52:14give your workspace an identity and

52:16allow it to connect to ADLs Gen 2 create

52:20your shortcuts things like that in a

52:22secure manner kind of trusted workspace

52:24access is what it's called and if you

52:26want to learn more more about that I'll

52:27leave a link to the workspace identity

52:29section in the school Community we can

52:31also do things like adding private

52:33endpoints and that kind of thing for

52:34connecting via spark to things in Azure

52:37you've got your spark settings down here

52:40for configuring something about the pool

52:43that you're using the spark pool that is

52:45being used in this workspace you can

52:47change the default environment that's

52:49where you're going to be going to add

52:51libraries and things like that so if you

52:53want to pre-install python packages onto

52:55your spark cluster so that every time

52:58you run a notebook or start a new

53:00notebook you have those libraries there

53:02ready to go that's where you do this

53:03change some settings for high con

53:05currency and that kind of thing there so

53:07Camila says okay great I now have some

53:09clarity on administering fabric at the

53:12tenant capacity in the workspace level

53:14what I'm not sure about is giving access

53:16to the people on my team so that's

53:18important right we build all these

53:20things in fabric but how can we give

53:22people access the right amount of access

53:25to these items that's a look at that in

53:27a bit more detail so if we go back to

53:28our structure of a fabric implementation

53:32generally when we're sharing items with

53:35people that's really done at these

53:37bottom three levels of our fabric

53:40architecture right so sharing things in

53:43fabric is normally done at these three

53:45levels now object level sharing is

53:48possible for the data warehouse and the

53:50seal endpoint in the lakeh house but I

53:52don't think it's assessed as part of

53:53this exam so it's not part of the study

53:56guide in in anyway so we're not going to

53:57be covering that in this lesson there's

53:59some documentation on the Microsoft

54:01learn website and I'll leave a link to

54:02that if you're interested in object

54:04level sharing if you want to bit learn a

54:06bit more about that we're going to be

54:08focusing on the workspace level sharing

54:10and item level sharing so let's just

54:12start with workspace level sharing

54:15people or groups can be given workspace

54:18level access and when sharing the

54:21personal group is assigned a workspace

54:23role as you can see on the right hand

54:25side there we've got admin member

54:27contributor and viewer now this role

54:30applies to all items in the workspace

54:33for example a viewer in the workspace

54:35will be able to view all of the items in

54:37the workspace let's just take a bit of a

54:39moment cuz roles are really important

54:42and the role that you assign someone

54:44dictates basically what they can do in

54:46your workspace this image here comes

54:48from the Microsoft documentation again

54:51I'll leave a link to this in the school

54:53community and I definitely recommend you

54:55take some time to study it this is what

54:57we're going to be doing here so the

54:58first thing to note with this diagram is

55:01let's just start with the admin so if

55:03you give someone an admin permission

55:05what can they do the first thing is that

55:06they can update and delete the workspace

55:08so this is really high level permissions

55:11that only maybe one or two people really

55:13should have people that you trust in

55:14your organization they can also add and

55:17remove people including other admins so

55:20it's the only role that allows you to

55:22add an admin add another admin next we

55:24move down to the member and the member

55:27can do similar things to an admin but

55:30they can't add an admin okay so a member

55:33cannot add an admin they can only add

55:35people with lower permissions or other

55:38members okay the other permission that

55:40is unique at the member level is you can

55:43give other people the permission to

55:45share items so being able to share items

55:48is a fairly high level thing to do and

55:50you're giving people that permission to

55:51share okay so that's something to bear

55:54in mind as well then we move down to the

55:56contributor level now contributors can

55:59do pretty much everything in the

56:01workspace other than as we see here

56:03deleting the workspace adding other

56:06people into the workspace and allowing

56:08other people to share but they can do

56:09when you're talking about contributing

56:12to fabric items anything around lake

56:15houses or warehouses or data pipelines

56:18they have read and write access to all

56:20of these things so the viewer has a

56:22unique set of missions those six green

56:25ticks

56:26and if we go kind of from top to bottom

56:28they can view and read content in a data

56:31pipeline a notebook spark job definition

56:33machine learning model so they can view

56:35kind of the outputs of these things they

56:37can also View and read the content of

56:39kql databases query sets and real time

56:42dashboards they can connect to the SQL

56:44analytics endpoint of a Lakehouse or a

56:47data warehouse and they can read

56:49Lakehouse and data warehouse data and

56:51shortcuts with tsql so the viewer can

56:54basically use SQL to analyze data in

56:57either The Lakehouse or the data

57:00warehouse what they can't do is access

57:03any of the one lake apis or spark so

57:06they can't run spark jobs or notebooks

57:10or anything like that now one unique

57:11thing about the viewer permission is in

57:14the data pipelines now they can't edit

57:16or update any of the activities in a

57:19data pipeline but they can execute and

57:23cancel the execution of a data pipeline

57:25run so that's an important kind of edge

57:28case to remember for the exam and they

57:30can also finally view the output of data

57:33pipelines notebooks and machine learning

57:35models so that's kind of a high level

57:37overview of all of these workspace roles

57:40and what they can do at each level again

57:42this is really important to understand

57:44for the exam so I definitely recommend

57:46going into the documentation taking some

57:48time to understand these different

57:49things because you'll probably be tested

57:51quite a lot on these so let's just have

57:52a bit of a workspace level access

57:55example this is John he is a business

57:58analyst working in Camila's team and

58:01Camila has asked you to give him

58:03contributor access to workspace one this

58:06is workspace one this is the

58:08architecture that they've got here now

58:09this is what John's access currently

58:11looks like where the red box is

58:13basically no access at all and a green

58:15box if there's any access and you can

58:17see everything's red So currently has no

58:18access to anything you are an admin in

58:21the workspace now what steps would you

58:23take to give John this access have a

58:25look little think about that and then

58:27we'll talk about it in a second okay so

58:29what steps would you take well

58:31personally what I would be asking is

58:33does John fit into an existing security

58:36group that has contributed access to the

58:39workspace because best practice here is

58:41to add people into groups rather than

58:43adding them individually just makes

58:44maintenance in the future a lot easier

58:47where possible we always want to add

58:49people into groups before we add them

58:51individually now if a security group

58:54doesn't exist then you might want to

58:55create one for John maybe you want to

58:57create an analyst security group so that

58:59in the future when another analyst wants

59:01to join the team or join the workpace

59:04you can just add that person into the

59:06group rather than having this long list

59:09of individual contributors in that

59:11workspace so you create an analyst

59:12Security Group add John to the security

59:15group and give the group contributor

59:17access to workspace one so in the

59:19picture how does that update well it

59:21looks a bit like this right so John now

59:23has access to workspace one and

59:26everything within it because we've given

59:28him access the security group access at

59:31that workspace level you'll notice that

59:33workspace 2 he still has no access to

59:36that he can't even see that so that's

59:37something to bear in mind when you're

59:39giving workspace level access Camila

59:42suddenly Rings you she realizes that JN

59:44shouldn't have access to everything in

59:46the workspace instead she wants you to

59:49give him access to the data warehouse

59:51only not the semantic model not the data

59:54pipeline so how would you check change

59:56what we've just done to reflect this so

59:59this is what we're we're looking at here

1:00:00we want to go from this which is the

1:00:02arrangement that we've just done for

1:00:04John at the workspace level to this at

1:00:07the item level now this might be

1:00:08important because it kind of reflects

1:00:11quite an important principle when it

1:00:13comes to giving people access which is

1:00:15the principle of least privilege now in

1:00:18general in Data Systems information

1:00:21security we want to give people the

1:00:23amount of access that they need to

1:00:25perform their roles and nothing more

1:00:27right so if you don't technically need

1:00:29access to the semantic model or the data

1:00:31pipeline then one way of kind of getting

1:00:34around that is to give people item level

1:00:36access giving people access to only what

1:00:38they

1:00:40need okay just to recap on some of the

1:00:44additional permissions so when you share

1:00:46a data warehouse you get these three

1:00:49additional permissions we have read all

1:00:51data using SQL and what that means is it

1:00:55allows people to read all objects within

1:00:57the warehouse using tsql we also have

1:01:00the read all one L data and with this

1:01:03you're allowing that person to read the

1:01:05underlying oneel files using spark

1:01:08pipelines anything else basically so in

1:01:10the top one they can only use SQL if you

1:01:13give them the second permission it

1:01:14allows them to basically do anything

1:01:15with that data and the third permission

1:01:18allows the user to build reports on the

1:01:20default semantic model not any custom

1:01:22models just the default semantic model

1:01:24when it comes to the lak house these

1:01:25permissions are similar but they're just

1:01:27worded a little bit differently so again

1:01:29if you give them the read all SQL

1:01:31endpoint data it allows them to perform

1:01:34tsql on the tsql endpoint if you give

1:01:37them read all Apache spark then again

1:01:41it's going to allow them to run

1:01:42notebooks and Spark code on top of that

1:01:44data and again the build reports on the

1:01:47default semantic model does exactly what

1:01:49it says on the tin now one point that I

1:01:51just did want to make here is around one

1:01:53Lake data access model now this is a

1:01:56very newly announced feature so it might

1:01:59not have made its way into the exam yet

1:02:01but I do think it's going to have a very

1:02:02big impact on how we manage security in

1:02:06fabric going forward so I did want to at

1:02:08least mention it here I'm not going to

1:02:10be going through it in detail I did just

1:02:12want to flag it you might want to have a

1:02:13look at the documentation page just so

1:02:16that you become aware of it now this

1:02:18feature is not really something I looked

1:02:20at yet much in detail but from just from

1:02:22looking at the documentation from how I

1:02:24understand it is it's going to allow you

1:02:25to perform rback so Ro based access

1:02:29control on things like folders so now

1:02:32that they've implemented folders within

1:02:33a workspace it's going to allow you to

1:02:35define a specific role or give a

1:02:38specific role access control over that

1:02:41folder and then the permissions are

1:02:42going to be inherited for every item in

1:02:44that folder but like I mentioned it's a

1:02:46preview feature and it's relatively new

1:02:48so I'd be surprised if they ask you

1:02:50about this in the exam but it is very

1:02:52important I do think it will change

1:02:53quite a lot in fabric so I wanted to

1:02:55mention it and if you look at the study

1:02:57guide it does say for the dp600 exam it

1:03:00does say that most questions cover

1:03:02features that are generally available

1:03:03the exam may contain questions on

1:03:05preview features if those features are

1:03:07commonly used I think at the moment this

1:03:09isn't commonly used because it's only

1:03:10been released a few weeks ago so that's

1:03:12something to bear in mind Camila says

1:03:14thank you now I understand workspace

1:03:16level and item level sharing in more

1:03:18detail One Last Thing Before You Go

1:03:20we've been working on this government

1:03:22project and I need to apply sensitivity

1:03:24labeling in a workspace can you walk me

1:03:27through it so what even is a sensitivity

1:03:29label well sensitivity labels are a data

1:03:32governance feature and they're created

1:03:35and managed in Microsoft purview so

1:03:38fabric items such as a semantic model

1:03:40can be given a sensitivity label such as

1:03:43confidential right and it's for

1:03:45information protection purposes now in

1:03:47some Industries labeling data and

1:03:49information with a sensitivity label is

1:03:52necessary for compliance with

1:03:54information protection regulations now

1:03:56to apply a sensitivity label in fabric

1:03:59really there's two main methods if we go

1:04:02into that item for example this

1:04:04Lakehouse here what you have in the top

1:04:06tool bar you've got the sensitivity

1:04:08label and you can just click on that

1:04:09drop- down change sensitivity label in

1:04:12there the other option is to go into the

1:04:15settings of that particular fabric item

1:04:17and you can see that in the left hand

1:04:19tool bar there you've got sensitivity

1:04:21label and you can change the sensitivity

1:04:22label in there now one of the options

1:04:25that you can give it if you go through

1:04:26the settings method is to apply to

1:04:28Downstream items so again we've got that

1:04:31notion or that concept of inheritance of

1:04:34the label that you give it here also

1:04:36applies to everything Downstream okay so

1:04:38now we are going to test some of your

1:04:39knowledge for everything that we've

1:04:41learned in this section of the study

1:04:44guide and we're going to start with a

1:04:46case study style question so we're going

1:04:48to going through a bit of a case study

1:04:50and then going to be asked three maybe

1:04:52four questions on this particular case

1:04:54study let's again Toby creates a new

1:04:57workspace with some fabric items to be

1:04:59used by data analysts Toby creates a new

1:05:03security group called Data analysts he

1:05:05includes himself as a member of this

1:05:07Security Group Toby gives the data

1:05:09analyst Security Group a viewer role in

1:05:13the workspace what workspace role does

1:05:15Toby have is it a viewer B member C

1:05:18admin or D contributor pause the video

1:05:21here have a think and then we'll move

1:05:23forward to the answer so the answer here

1:05:25is see now this combines two pretty

1:05:27important Concepts to understand when

1:05:29we're looking at workspace level sharing

1:05:32number one is that the creator of a

1:05:35workspace is always given admin

1:05:37permissions in that workspace now we

1:05:39also have Toby with the viewer role in

1:05:41that workspace CU he's in the security

1:05:43group with viewer role and this is

1:05:46another concept if you have more than

1:05:47one level of permission within the

1:05:49workspace you're always given the higher

1:05:51level so he's got admin role cuz he

1:05:53created the workspace and he's got

1:05:55viewer role because he's in that

1:05:56Security Group Well the admin

1:05:58permissions is always going to be

1:05:59prioritized he's always going to take

1:06:01that role over his viewer role so let's

1:06:03continue this case study Sarah is also a

1:06:07member of that data analyst Security

1:06:09Group she has no other role in the

1:06:11workspace which of the following can

1:06:13Sarah not do in the workspace a execute

1:06:16a data pipeline run SQL scripts in the

1:06:19data warehouse run spark notebook or

1:06:22review the evaluation metrics of a

1:06:24machine learning model now the answer

1:06:25here is C run the spark notebook now

1:06:28when we're looking at the workspace

1:06:30level roles and the permissions for each

1:06:32role we know that Sarah is a viewer in

1:06:34the workspace that's the highest level

1:06:36of permission and you remember that a

1:06:38viewer role can actually execute a data

1:06:41pipeline in a workspace they can also

1:06:44run tsql scripts in data warehouse or a

1:06:47SQL endpoint of a lake house what they

1:06:49can't do is run a spark notebook okay so

1:06:52anything in a notebook they can have a

1:06:54look at the notebook but they can't

1:06:56actually execute any code so C is the

1:06:58right answer cuz what we're looking for

1:07:00is what can she not do in the workspace

1:07:03and D is review the evaluation metrics

1:07:05of a machine learning model which we

1:07:07know we can do because she's just

1:07:08reading the output of that model to

1:07:11continue this case study again Toby

1:07:13wants to delegate some of the management

1:07:15responsibility in the workspace he wants

1:07:17to give this person the ability to share

1:07:20content within the workspace invite new

1:07:23contributors to the workspace but not

1:07:25add new admins to the workspace what

1:07:27role should Toby give this person a

1:07:29admin B member C contributor or D viewer

1:07:33so the answer here is B now the the key

1:07:36point in the question was but not add

1:07:38new admins to the workspace so we know

1:07:41that to be able to add another admin

1:07:42into a workspace you need to have admin

1:07:44permissions yourself so Toby doesn't

1:07:46want to give that person this ability

1:07:48basically so we know it can't be admin

1:07:50it's not going to be viewer it's not

1:07:52going to be contributor we know that the

1:07:54member is kind of one down from admin

1:07:57and that's going to allow you to do all

1:07:58of these three things they can share

1:08:00content they can invite other

1:08:02contributors because a member can add

1:08:04new people either members or

1:08:06contributors or viewers but they can't

1:08:08add other admins so it' be B member the

1:08:11next question is completely separate you

1:08:13have admin role in a workspace Sheila is

1:08:16a data engineer in your team she

1:08:18currently has no access to this

1:08:20workspace at all now Sheila needs to

1:08:22update a data transformation script in a

1:08:24pisb notebook and the script gets data

1:08:27from a Lakehouse table cleans it and

1:08:29then writes it to a table in the same

1:08:30Lakehouse now you want to adhere to the

1:08:33principle of leas privilege what actions

1:08:35should you take to enable this is it a

1:08:38you're going to give Sheila the

1:08:39contributor role in the workspace b

1:08:42share the Lakehouse item with read or

1:08:44spark data permission C give Sheila the

1:08:47admin role in the workspace or D share

1:08:50the lake house item with read all spark

1:08:52data permissions and share the notebook

1:08:54with edit permissions so the answer here

1:08:56is D so one of the clues in this

1:08:58question was the line where it says you

1:09:01want to adhere to the principle of leas

1:09:02privilege so immediately when you see

1:09:04that giving people workspace level

1:09:07access is not really good enough so A

1:09:09and C is giving a role in the workspace

1:09:13so it's going to enable her to

1:09:14contribute and change and edit

1:09:16everything in the workspace but it

1:09:18doesn't adhere to the principal of least

1:09:19privilege so we can immediately rule out

1:09:21a and C so another really important

1:09:25point in the question here was Sheila

1:09:27needs to update a data transformation

1:09:30script in a notebook so she needs to

1:09:32edit the code in a notebook She's Not

1:09:34Just executing an existing notebook she

1:09:36needs to actually make changes to a

1:09:38notebook and so for B you wouldn't have

1:09:41that permission you've got read all data

1:09:43for the spark so you can actually

1:09:45execute a notebook but you can't edit a

1:09:47notebook so be able to make these

1:09:49changes really need access to the

1:09:51notebook and The Lakehouse that that

1:09:54notebook is interfacing with that that

1:09:56notebook is reading from because you

1:09:58can't just share the notebook because

1:09:59then you won't have access to the

1:10:00underlying data and we can't just share

1:10:02the lake housee because you won't have

1:10:03access to the notebook that she needs to

1:10:05edit so the answer is D share the

1:10:07Lakehouse item we're giving spark

1:10:09permissions and we're also giving edit

1:10:11permissions on the notebook next

1:10:13question you have admin role in a

1:10:15workspace you want to pre-install some

1:10:17useful python packages to be used across

1:10:20all notebooks in the workspace how do

1:10:22you achieve this a in the fabric ad

1:10:25admin portal go to spark settings and

1:10:27install the libraries B go to workspace

1:10:29settings spark settings and then Library

1:10:32management C create an environment

1:10:34install the packages in the environment

1:10:36go to the workspace settings spark

1:10:38settings and set it as the default

1:10:40environment or D go to capacity settings

1:10:43and then default libraries so the answer

1:10:45here is C creating an environment and

1:10:49then going into your workspace settings

1:10:50and setting it as the default

1:10:53environment for for spark now now this

1:10:55is a bit of a a naughty question because

1:10:57B is the old way so it used to be you go

1:11:00to workspace settings spark settings and

1:11:02there was a section for Library

1:11:03management but that's actually not

1:11:05possible anymore the way to do it as I

1:11:07mentioned is to create an environment

1:11:09and then in your spark settings make it

1:11:10the default environment A and D don't

1:11:13actually exist these capabilities so

1:11:16these kind of red herrings so the answer

1:11:18is C Camila says thanks she's seriously

1:11:20impressed with your knowledge again in

1:11:22this lesson we covered all five of these

1:11:25elements of the dp600 study guide from

1:11:28workspace and item level sharing data

1:11:30sharing for data warehouses and lake

1:11:33houses sensitivity labeling and then

1:11:36workspace and capacity level settings

1:11:39and the good news is again you've won an

1:11:40extension to the contract Camila would

1:11:42like you to implement control over the

1:11:45entire analytics development life cycle

1:11:48in her organization so for this we're

1:11:50talking Version Control deployment

1:11:52pipelines powerbi projects all that good

1:11:55stuff that's what we going be looking at

1:11:56in the next lesson so click here to

1:11:59continue that lesson hey everyone

1:12:01welcome back to the channel today we're

1:12:03continuing the DP 600 series looking at

1:12:06what it's going to take to hopefully P

1:12:08that exam and become a fabric certified

1:12:11analytics engineer Today Is video 4

1:12:14we're going to be looking at managing

1:12:16the analytics development life cycle and

1:12:18in this section exam we're going to be

1:12:20focusing on implementing Version Control

1:12:23creating and managing powerbi projects

1:12:26planning and implementing deployment

1:12:27Solutions performing impact analysis on

1:12:31Downstream activities deploying and

1:12:33managing semantic models through the

1:12:35xmla endpoint and creating reusable

1:12:38assets so powerbi template files powerbi

1:12:41data source files all these kind of

1:12:42things so that's what we've got in store

1:12:44for you today as ever there's going to

1:12:46be some sample questions at the end I'll

1:12:48also be posting all of the lesson notes

1:12:52and links for further resources in the

1:12:54school community go and grab them there

1:12:56you will play the main character in the

1:12:58scenario again we're going to be

1:12:59continuing this theme this scenario that

1:13:02we've been developing throughout this

1:13:03course you're the main character are you

1:13:05ready let's begin so as you know already

1:13:07we are a consultant working with the

1:13:10client called Camila and you've already

1:13:12helped Camila plan and Implement her

1:13:14fabric implementation her environment

1:13:17now the time has come to implement

1:13:19analytics development life cycle now she

1:13:21wants to focus on Version Control at

1:13:23least initially but to be honest he's

1:13:25not really familiar with Git Version

1:13:28Control never really used that in an

1:13:29analytics environment before so what

1:13:31you're going to have to do is set up a

1:13:34call with Camila and walk her through

1:13:36the basics of git first and Version

1:13:38Control what is it why does it even

1:13:40exist why is it now a part of fabric and

1:13:44what are some of the key terms and

1:13:45terminology that we need to understand

1:13:47when we're implementing Version Control

1:13:49just bear in mind that all of the git

1:13:52integration features are currently in

1:13:54preview and some of the other features

1:13:56we're going to be looking at today like

1:13:57the deployment pipeline functionality a

1:14:00lot in this kind of analytics life cycle

1:14:02stuff is still in the previous stages so

1:14:04just bear that in mind when we're

1:14:06walking through these examples I'll show

1:14:07you what's possible today but if you're

1:14:09coming from using like GitHub or git a

1:14:12lot in the past and a lot of the

1:14:14features that you might be used to are

1:14:16currently not available so let's move on

1:14:19okay so I just wanted to start this

1:14:21video with a bit of a primer on git and

1:14:24Version Control in general and I think

1:14:26the best way to do that is to show you

1:14:28what this looks like and along the way

1:14:30we can explain some of the key Concepts

1:14:32in git and Version Control I realize

1:14:35that probably a lot of people who have

1:14:36been coming from maybe a powerbi

1:14:38background don't have experience with

1:14:40Version Control and get in general so

1:14:42let's set up a project and start at a

1:14:45basic level introduce more and more

1:14:47Concepts as we go through this demo so

1:14:50in fabric the way you're going to start

1:14:52is with Azure devops now AZ devops is a

1:14:55service that's used primarily for

1:14:57software development and managing

1:14:59infrastructure and the devops life cycle

1:15:02of that infrastructure and software

1:15:04development but it's also where we're

1:15:06going to use to store our code and our

1:15:09artifacts and our powerbi project files

1:15:12because they come with repositories so

1:15:14you need to set up an account just go to

1:15:16this website here and I'll leave a link

1:15:18in the description or in the school

1:15:19Community now once you've set up Azure

1:15:21devops for your organization you'll come

1:15:24through to this page here I've set up an

1:15:27organization called fabric University

1:15:29and then you have this concept of a

1:15:31project so a project is going to be

1:15:33where you store your repositories your

1:15:35code has lots of other features as well

1:15:37but we won't be going into too many of

1:15:39the other features in this short video

1:15:41so we're going to be creating a new

1:15:42project for this demo let's just call it

1:15:45dp600 practice we're going to make it

1:15:48private and then just click on create

1:15:50okay so this is an Azure devops project

1:15:53on the left hand side you can see here

1:15:55make that a little bit bigger on the

1:15:56left hand side we've got some useful

1:15:59things to know about right so this is

1:16:01just the overview where you can see an

1:16:02overview of the project probably the

1:16:04most important ones to bear in mind are

1:16:06boards so this is like a Work Management

1:16:08tool for doing development work we're

1:16:11not going to be focusing much on that in

1:16:12this video what we're interested in is

1:16:14this Repose repositories right so

1:16:17repositories is kind of the core thing

1:16:19that we need to implement Version

1:16:22Control now repositories are also

1:16:24available in GitHub but currently that

1:16:26integration is not possible between

1:16:28Fabric and GitHub you can only use Azure

1:16:31devops repos so that's why we're using

1:16:34this currently so a repository is just

1:16:36where we're going to store our code in

1:16:39the cloud so when you're developing

1:16:41locally like a powerbi report for

1:16:43example really what we want to do is

1:16:46store that report in the cloud in our

1:16:50repository and that's where we're going

1:16:51to be controlling tracking changes

1:16:54managing who can make changes to that

1:16:56project in the repository so let's just

1:16:59do a bit of configuration here to set up

1:17:00this repository so that we can use it

1:17:02for Version Control so the first thing

1:17:04I'm going to do is right at the bottom

1:17:06here initialize a main branch with a

1:17:08readme and a readme file is just a

1:17:11markdown file it's like a text file just

1:17:13to initialize our branch and we'll talk

1:17:16a bit more about branches in a bit more

1:17:18detail here currently I have one branch

1:17:20and it's the main branch here now

1:17:22branching is kind of like a whole field

1:17:24in itself there's lots of different

1:17:26strategies for how we can manage

1:17:28different branches when we developing

1:17:30our code or developing our fabric

1:17:33artifacts but we'll get into that in a

1:17:34bit more detail a bit later on for now

1:17:37what we're interested in really is

1:17:39creating a copy of this repository on

1:17:43our local machine right so that's going

1:17:45to be the first sync we want to get

1:17:47these two in sync because the way that

1:17:50version control works in general is you

1:17:53develop locally on your local machine

1:17:55for powerbi and then you sync those

1:17:57changes up to the repository in the

1:18:00cloud and if you have a big team of

1:18:02people maybe you have 10 people doing

1:18:03this they're all syncing their changes

1:18:06into this main branch by default so what

1:18:09we want to do is clone this repository

1:18:12and we get this URL here for this git

1:18:14repository we can copy the URL and what

1:18:17we want to do is clone that on our local

1:18:19machine so there's various different

1:18:20ways to do that you might see in the

1:18:22documentation Microsoft use a tool

1:18:25called vs code I personally prefer using

1:18:28GitHub desktop this is another tool it's

1:18:30completely free you can download it just

1:18:33Google GitHub desktop download it's

1:18:35built by obviously GitHub but it's also

1:18:38possible to do Azure devops repos in

1:18:41here as well and it just makes the whole

1:18:43process of Version Control and git a lot

1:18:47simpler it basically builds a UI on top

1:18:49of a lot of the git functionality so the

1:18:52first thing we want to do is click on

1:18:53file and clone repository because we're

1:18:56trying to drag things down from the

1:18:59internet and if you click over to URL we

1:19:01can just copy the URL that we've got

1:19:03here that we got from our azid Devo

1:19:05repository and we can clone it now it's

1:19:07going to ask for a username and password

1:19:09so if you go back to a devops we've got

1:19:12generate this git credentials and we can

1:19:15just copy the username in there copy the

1:19:18password in there save and retry so

1:19:20that's just going to authenticate with

1:19:22our Azure devops account so we know that

1:19:24okay this is the actual person that owns

1:19:26that repository it's creating that

1:19:28authentication between the two okay so

1:19:30what has that actually done well now

1:19:32what we've got is a copy of what we have

1:19:35in Azure devops in our local machine you

1:19:38can see the file path there is C users

1:19:41learn documents git and what we can

1:19:44actually do is open this in Explorer

1:19:47okay so now I have a copy of those files

1:19:50and folders from Azure devops on my

1:19:53local machine so that is the first first

1:19:54thing that we need to do really is clone

1:19:57the repository get that copy local and

1:19:59what's going to happen now we've got

1:20:01this all set up and synced so that

1:20:04GitHub desktop is Now tracking this

1:20:06folder so any changes you make locally

1:20:09to this folder it's going to pick those

1:20:11up right so for example if I do a new

1:20:14text file call it my file now if we go

1:20:16back to get up desktop you can see that

1:20:18it's picked up that change already it's

1:20:20saying you've added a new file in there

1:20:23it's called my file . text and if I open

1:20:26that with notepad for example just call

1:20:28it my amazing file and I'll save that

1:20:31here we can see again it's tracked to

1:20:32the changes so git works by tracking the

1:20:35changes in text based documents okay so

1:20:40traditionally it's used for code so a

1:20:43python file a SQL script or a Javascript

1:20:46file or C all these kind of things are

1:20:49textural based right so they work really

1:20:52well with Git so now we've made some

1:20:54Chang changes to our local repository

1:20:57right we've added this text file but

1:20:59what you'll notice is that you know here

1:21:01in Azure devops nothing has changed yet

1:21:03because what we need to do is push those

1:21:05changes into the cloud environment maybe

1:21:09you've got a team you want other people

1:21:10to see the changes that you've made and

1:21:12we'll start by using this text file

1:21:15right and then we'll slowly move up to

1:21:16more complicated stuff we'll look at

1:21:18powerbi projects as well but for now

1:21:21let's just push these changes so you see

1:21:23down here you can do created a new file

1:21:26so created a new file and then what

1:21:28we're going to do is click this commit

1:21:29to main button so this is going to

1:21:32commit our changes onto that main branch

1:21:35and up here we click on push to origin

1:21:38so we're going to commit and then push

1:21:40the changes up into the cloud right into

1:21:42Azure devops and so now what's going to

1:21:44happen if we refresh our Azure devops we

1:21:47can see that file there we can see my

1:21:49file. text my amazing file that's in

1:21:52there in this repository in azure devops

1:21:55now Okay then if we want to make more

1:21:57changes maybe we want to change this a

1:21:59bit more even more amazing file and we

1:22:01save that again it's going to show you

1:22:03oh well your original file was this okay

1:22:06this is what we've got stored in Azure

1:22:08devops this is what the the main branch

1:22:11is currently saying now you've changed

1:22:13that file right and we've noticed that

1:22:15okay so it's tracking the changes

1:22:17between every change you make in that

1:22:19file so that's really important thing to

1:22:21bear in mind so let's just update this

1:22:24text file obviously in practice in

1:22:26proper environments you add a more

1:22:28meaningful commit message because these

1:22:29are really important and we'll push that

1:22:31again to the origin using this button

1:22:33here so again if we just go back to the

1:22:36Azure devops you can see that that has

1:22:38now updated in here as well great so

1:22:40that is the very very basic

1:22:42implementation of git and Version

1:22:45Control and syncing our local repository

1:22:48with our Azure devops repository in the

1:22:50cloud so at this stage I think it's

1:22:52worthwhile just noting what we we've

1:22:54done here right so at the top we've

1:22:56created an A devops account we've added

1:22:59that readme file we've copied the git

1:23:01URL and cloned it locally and then we've

1:23:04modified that local folder create a new

1:23:06file modified the file committed those

1:23:09changes back pushed the origin and

1:23:11observe the changes in as devops now

1:23:13this is great it's a good first step

1:23:14right we can track changes now between

1:23:16files but one of the core benefits of

1:23:19git and Version Control in general is

1:23:21that instead of just allowing anyone to

1:23:23update that main branch what we can do

1:23:26is we can protect that main branch so

1:23:29that we can control who has access to

1:23:32and who has the ability to overwrite

1:23:35those changes so let's make a few

1:23:37changes to the figuration of our

1:23:40repository in Azure devops what we're

1:23:42going to do is we're going to protect

1:23:44that main branch so that nobody can just

1:23:46overwrite that file anymore we want to

1:23:49add in a review and approval phase and

1:23:52this is going to help a lot with our

1:23:54quality and control over who and how

1:23:58that code base is being changed so let's

1:24:00have a look at that in a bit more detail

1:24:02okay so in our Azure devops project here

1:24:05we can go to Project settings and then

1:24:07we can go down to repositories and then

1:24:09what we're going to do is add in a

1:24:12policy for branch policies protect

1:24:15important branches Nam spaces so we're

1:24:17going to add something in here protect

1:24:19the default Branch create automatically

1:24:21included reviewers so I'm going to add

1:24:23myself in here as a reviewer just

1:24:26because there's only one person in this

1:24:28project in reality you might have a team

1:24:30of maybe senior developers or lead

1:24:32developers that responsible for

1:24:35reviewing code and approving changes to

1:24:38the the code base so now we've set that

1:24:40up let's have a look at doing something

1:24:42a bit more so we've got our file here so

1:24:45if I save this I've just added another

1:24:47line in here new changes and if we go

1:24:49back to our GitHub here we can see that

1:24:52oh it's picked up that new changes again

1:24:54but now what I'm going to do is update

1:24:58this so added some new changes and I'm

1:25:01going to commit that and when I try to

1:25:02push this to the origin is going to say

1:25:04error okay because now pushes to this

1:25:07Branch are not permitted you must use a

1:25:09poll request to update this Branch so

1:25:11that's going to introduce a few new

1:25:13Concepts that we need to get around this

1:25:17really number one is the concept of

1:25:19branches and number two is the concept

1:25:21of pool requests let's have a look at

1:25:23both of those now now so now that we've

1:25:25implemented some protection over that

1:25:28main branch we have to change how we

1:25:30develop okay so now we have to use

1:25:34branching So currently all of my files

1:25:37that text file and the readme file is on

1:25:40this main branch so if I now want to

1:25:42update what's in that text file I'm

1:25:45going to have to create a new Branch

1:25:46because nobody can update and edit that

1:25:49main branch directly so what we can do

1:25:51in GitHub desktop it's very easy so what

1:25:54we're going to do is click on Branch New

1:25:57Branch now we can change the name to

1:25:59change text file create this branch and

1:26:02now if we publish this branch and we can

1:26:03add in some changes here so now if we go

1:26:06back to Azure devops let's just go back

1:26:08to our project here and back to Repose

1:26:11and if we go down to this change text

1:26:14file you can see that it has actually

1:26:16found this new Branch this is a branch

1:26:19and if we go over to the pull requests

1:26:21section we'll explain what that means in

1:26:22a minute but you can see that it's

1:26:24registered the fact that we have now

1:26:26published a new Branch so we've made

1:26:29some changes added that new changes line

1:26:31in the text file and it's popped up

1:26:33saying oh do you want to create a pool

1:26:35request here and so the pool request is

1:26:37when we want to make a change or merge

1:26:40some changes into the main branch okay

1:26:43and you can only do this through a

1:26:45review because that's the policy that

1:26:46I've set up on that main branch so what

1:26:49we can do is create a PO request you can

1:26:51say oh I've added some new changes in

1:26:53here you can add in a reviewer which is

1:26:55going to be me and you can say oh please

1:26:57review this and you can create a pull

1:27:00request so what that's going to do is

1:27:01going to notify the reviewer and say oh

1:27:05this person wants you to review their

1:27:07changes to the code base on this change

1:27:09text file branch and now as the reviewer

1:27:12because I'm kind of playing both roles

1:27:14here I can have a look at the file

1:27:16changes I can review the changes like oh

1:27:19yep added new changes that looks good to

1:27:21me I'm going to approve that change and

1:27:23now what you can do is complete so what

1:27:25complete is going to do is merge those

1:27:29changes into the main branch so your

1:27:31protected Branch because we've gone

1:27:33through the reviewer process the text

1:27:36file has been reviewed now we can merge

1:27:38it safely into the main branch there's a

1:27:41few options here that are quite

1:27:43important to bear in mind and we'll look

1:27:46at those when we move into fabric we'll

1:27:48look at those in a bit more detail but

1:27:50generally kind of the default settings

1:27:52here complete Associated work items

1:27:54after merging it's fine we don't want to

1:27:57delete change to text file after merging

1:28:01generally in Fabric and we'll get to why

1:28:03that is in a minute we'll just do on

1:28:05complete merge Okay so we've now merged

1:28:07our poll request let's have a look at

1:28:10the repository so now we have my text

1:28:13file in the main branch and it's got new

1:28:15changes Okay so we've merged our changes

1:28:18into this main branch great that's the

1:28:21first section of this demo done okay and

1:28:24this second half there what we did was

1:28:25protect the main branch Okay so we've

1:28:28added an approv in the repository we've

1:28:31updated the repo approv policy we tried

1:28:35committing a new change and it was

1:28:36rejected we've added a new feature

1:28:38Branch we brought the changes onto that

1:28:40Branch we've committed that branch and

1:28:43we've added a p request then our

1:28:45approver has approved it and we've

1:28:47merged it into the main branch and just

1:28:49to reiterate the benefits of that is

1:28:51that now we're checking the changes

1:28:53between files and we're also controlling

1:28:55who can make changes so we're protecting

1:28:58that main branch and saying okay if you

1:28:59want to update this it has to go through

1:29:01an approval process now up until this

1:29:03point we've just been using one text

1:29:06file actually git can track the changes

1:29:09between any text based file format text

1:29:13base that's the the problem that existed

1:29:16with PB files PX files kind of a bit of

1:29:19a black box right if you can't open a

1:29:22file in notepad and understand what's

1:29:24going on and it's going to struggle with

1:29:26Source control Version Control in git

1:29:29now that is one of the core reasons why

1:29:31Microsoft developed the powerbi project

1:29:35file format because it represents a

1:29:37powerbi project in text based files

1:29:41right so let's just have a look at an

1:29:43example of a pbip first and then we'll

1:29:46look at how we can integrate that into

1:29:48Version Control Systems okay so here we

1:29:49are in powerbi desktop and I've just got

1:29:51one of the sample reports here from

1:29:54Microsoft the competitive marketing

1:29:56analysis report now what we're going to

1:29:58do is this is just a PB file at the

1:30:00moment what we're going to do is save

1:30:03this as a pbip a powerbi project file

1:30:07then we're going to explore that and

1:30:08have a look at what that looks like so

1:30:10it's fairly easy to save your powerbi

1:30:13file as a powerbi project file so if we

1:30:16go to file and then save as and what we

1:30:20can do is we can find our git reposit so

1:30:24that's in document git dp600 practice

1:30:28that's the repository that we just

1:30:29cloned from Azure devops and it's

1:30:31currently empty for this kind of thing

1:30:33we can change the save as type to a

1:30:35powerbi project and we can just save it

1:30:37like that click on Save and now we have

1:30:39our powerbi project saved in pbip format

1:30:43so let's just have a look at what that

1:30:45looks like okay so this is our git

1:30:47repository dp600 practice and we've

1:30:50saved our powerbi project as a BP file

1:30:55it's actually a number of different

1:30:57files and folders right what we've got

1:30:59here is we've got the read me that was

1:31:01already in the repository and my file is

1:31:03that text file that we've been working

1:31:05with but now what we've got is two

1:31:06folders a git ignore file and this pbip

1:31:11file format so what it's done is it's

1:31:14basically decomposed our powerbi project

1:31:17into a series of files and folders that

1:31:20describe what's going on in that report

1:31:23Okay so let's move back to uh GitHub

1:31:26desktop here and what you'll notice is

1:31:29well the first thing is I've created a

1:31:30new Branch okay so now that we're going

1:31:32to be updating this pbip file we'll

1:31:36notice that all of the files are now

1:31:37listed here CU we created that pbip

1:31:40they're all text based they're all

1:31:42trackable using Version Control first

1:31:44commit of the pbip again we can commit

1:31:47that up into Azure devops push that to

1:31:51the origin in Azure devops now if we go

1:31:54through into azid devops again poll

1:31:56requests we can see that here we can

1:31:58create a new poll request and then we

1:32:00can commit that in there the reviewer

1:32:03will be myself again create that approve

1:32:05it complete merge so now what we've got

1:32:09is our powerbi report in our Azure

1:32:13devops environment and it's been

1:32:15approved and it's now trackable through

1:32:17Version Control now imagine I'm a

1:32:19powerbi developer and I want to make

1:32:21some changes to that report what is that

1:32:24going to look like well we're going to

1:32:25create a new Branch so anytime you're

1:32:27making changes to your powerbi file

1:32:29you're going to create a new Branch

1:32:31powerbi changes obviously you'd make it

1:32:33a bit more descriptive than that but

1:32:35that's good enough for this purpose here

1:32:37then we're going to go through and

1:32:40actually this isn't a marketing report

1:32:41it's sales and marketing so we're going

1:32:43to update the title here just as an

1:32:45example we're going to save that file

1:32:47and now this is the real power of

1:32:50Version Control for our power VI assets

1:32:53but also all of the stuff in fabric that

1:32:55we'll have a look in a minute any

1:32:56changes that we make to that powerbi

1:33:00file are now being tracked in our

1:33:03Version Control System okay and then

1:33:05when we're done we can push those

1:33:07changes up into Azure devops go through

1:33:10the review process merge them into the

1:33:13the main branch okay so you're probably

1:33:14thinking great but up until this point

1:33:17you haven't even mentioned fabric I

1:33:19thought this was a channel talking about

1:33:20fabric well we just built up the

1:33:21concepts of git vers control we've

1:33:24looked at the local case of powerbi

1:33:27report development but everything else

1:33:28in fabric happens in the fabric Cloud so

1:33:31now let's look at where fabric fits into

1:33:33all this okay how do we set up a

1:33:35repository in our workspace how do we

1:33:38link a repository to our workspace let's

1:33:40have a look at how fabric fits into this

1:33:42picture now okay so here we are in

1:33:44Microsoft fabric what I've done is I've

1:33:46set up a workpace what we're going to do

1:33:49is link it to our Azure devops

1:33:52repository our Azure devops project

1:33:54so to do this you will need to be a

1:33:56workspace administrator go to workspace

1:33:58settings so the first time that you

1:34:01click on get integration you'll need to

1:34:03actually sync it with your account I've

1:34:05already done that with mine and it's

1:34:07going to allow you to select the

1:34:08organization the project the repository

1:34:11the P600 practice and a specific branch

1:34:14that you might want to connect to now

1:34:15one thing I would say is that the way

1:34:17that we've done that protecting of the

1:34:19branching in the last section of the

1:34:22video what I found in practice is that

1:34:24this doesn't work particularly well

1:34:25currently with the way that git

1:34:27integration is set up in fabric but

1:34:29we'll have a look at what it looks like

1:34:30just click on connect and sync okay so

1:34:33now our first Sync has started and it's

1:34:36been done successfully what you'll

1:34:37notice is this Source control section

1:34:40here it's got zero in it so our source

1:34:42control so our things in AZ devops

1:34:45repository and what's in fabric is

1:34:47perfectly in sync currently and we can

1:34:49see that as well with the green sync and

1:34:52green synced here so now that we've got

1:34:54uh what's in our Azure devops repost

1:34:56synced with what's in fabric now we can

1:35:00add in fabric items into this mix right

1:35:03so we can create a new data pipeline for

1:35:06example create a data pipeline at the

1:35:08moment I'll just keep it as an empty one

1:35:09and we can just go back to our

1:35:11repository and we can see that it has

1:35:14been uncommitted right and we can see

1:35:16the source control button up here is now

1:35:19saying one so if we click on that we can

1:35:20see that we got one change here so comp

1:35:23compared to what's in our main branch in

1:35:26AZ devops it's noticed that we've got a

1:35:28change so it's similar to how we saw in

1:35:31GitHub desktop which tracks local

1:35:33changes this is tracking changes in

1:35:36fabric as well so what we can do is

1:35:40commit this change this data pipeline

1:35:42into our source control in AZ devops now

1:35:45to do this because we've got a

1:35:47protection on that main branch let's

1:35:49just try committing and see what it says

1:35:51first commit data pipeline but we know

1:35:53know that we've got protections on that

1:35:55main branch so when we commit I don't

1:35:57think it's going to allow us to do that

1:35:59no forbidden due to the branch policy

1:36:01and this is where it gets a bit more

1:36:02complicated within fabric what we're

1:36:04going to have to do is check out a new

1:36:06Branch so maybe we want to do adding

1:36:10data pipeline that could be our new

1:36:12branch and then that's going to allow us

1:36:14to commit added a new data pipeline

1:36:16we're going to click on the item that we

1:36:18want to merge which is our data pipeline

1:36:20or commit not merge and now that's going

1:36:22to go through into Azure devops okay so

1:36:26if we open up our poll requests section

1:36:28boom we can see adding data pipeline so

1:36:31now we've got these two things in sync

1:36:33between Fabric and Azure devops and our

1:36:36local repository for powerbi development

1:36:38and Azure devops and again if we want to

1:36:40merge that into the code base into the

1:36:43main branch we can go through a familiar

1:36:45process of choosing a reviewer creating

1:36:48that PLL request and if you're the

1:36:50approver you can then approve that PLL

1:36:52request and complete and merge it into

1:36:55the main branch so then if we go back to

1:36:57our repository now we can see that in

1:37:00our main branch we have this my data

1:37:01pipeline data pipeline so that is now

1:37:04being tracked with Version Control in

1:37:07our repository and if we go back to

1:37:09fabric we can say that we've now got

1:37:12three synced items here so it's

1:37:15perfectly in sync with what's in our

1:37:17Branch now the problem here or the

1:37:20slight limitation that I found is

1:37:22changing back to the main branch cuz

1:37:25currently we're on this adding data

1:37:27pipeline Branch okay and that was just a

1:37:30feature Branch we want to move back onto

1:37:32the main branch normally so that we can

1:37:34maybe create another another Branch but

1:37:35we don't want to be branching from this

1:37:37added data pipeline we want to be

1:37:39branching from the main branch now that

1:37:41is possible we can just go back to

1:37:43workspace settings go through to the get

1:37:46integration again and we can change our

1:37:48Branch back to the main so we're going

1:37:50to switch and override and now we're

1:37:51going back to the the main branch and

1:37:54all three items are synced because we

1:37:55know that we've merged into that main

1:37:57branch but the problem here is that only

1:37:59the workspace administrator can change

1:38:01that Branch back so if you found any

1:38:03better ways of working with protected

1:38:05branches in git and fabric please let me

1:38:08know okay so we've covered quite a lot

1:38:10of ground there from the basics of git

1:38:13and Version Control in general right

1:38:15through to syncing our changes or

1:38:18tracking changes in a powerbi project

1:38:20into AZ devops and then the same from a

1:38:23fabric environment so tracking changes

1:38:24that we make within fabric also to the

1:38:27same Azure devops repository let's just

1:38:29do a bit of a summary there and focus

1:38:31back on the exam because obviously not

1:38:33all of that is going to be tested some

1:38:34of it was just for your background

1:38:36knowledge so in general Version Control

1:38:38with Git allows you to track changes