Free Webinar: Exploring Microsoft Fabric

IT Training and Development Resources
Webinars
Exploring Microsoft Fabric: A Unified Data Engineering and Analytics Platform [Webinar]

Microsoft Fabric empowers your organization to transform raw data into actionable insights quickly and efficiently. This unified data platform streamlines your data processes by combining data engineering, data science, real-time analytics, and business intelligence - all in one place. In this free 1-hour webinar, Faheem Javed, experienced Microsoft Fabric practitioner and instructor, discusses how Microsoft Fabric can turn data into a strategic advantage, giving your organization the competitive edge it needs to succeed.

Microsoft Fabric's architecture and components
The key features and capabilities of Microsoft Fabric for data engineering, data science, and data analytics.
The differences between OneLake, Lakehouse, and Data warehouse
How to work with pipelines and workflows using Microsoft Fabric
Integration with Power BI to create reports and dashboards

At the conclusion, Faheem will eagerly take your questions.

Target Audience:

Data engineers, data analysts, IT professionals, and decision-makers seeking to optimize their data infrastructure and analytics capabilities using Microsoft Fabric.

About the presenter:

Faheem Javed began his IT career as a Visual C++ programmer in the 1990s. Faheem has closely followed Microsoft's Business Intelligence evolution, witnessing transitions from acquiring third-party tools like Proclarity Analytics to advancements in SSIS/SSAS/SSRS, SharePoint Performance Point Services, Polybase, Power BI, Azure Synapse, Azure Databricks, and Microsoft Fabric. He holds numerous certifications, including Azure Solutions Architect and Azure Data Engineer, and has conducted standard and custom training on these technologies. Faheem also teaches Microsoft Fabric courses with Axcel ILT

In this session, you will learn:

Microsoft Fabric's architecture and components
The key features and capabilities of Microsoft Fabric for data engineering, data science, and data analytics.
The differences between OneLake, Lakehouse, and Data warehouse
How to work with pipelines and workflows using Microsoft Fabric
Integration with Power BI to create reports and dashboards

Target Audience:

Data engineers, data analysts, IT professionals, and decision-makers seeking to optimize their data infrastructure and analytics capabilities using Microsoft Fabric.

About the presenter:

WEBVTT

1
00:00:03.630 --> 00:00:23.699
Axcel ILT: Hi, everybody! Welcome to this Thursday's webinar exploring Microsoft fabric. Thank you so much for being here. We know that you're busy with lots of other things to do. But I'm here with my colleague, Fahim javeed, and he's going to be presenting Microsoft fabric to you, and we will start in just a moment.

2
00:00:24.765 --> 00:00:43.050
Axcel ILT: I just wanted to welcome you on behalf of Excel instructor led training, and in case you didn't know, accelerate, exit certified, and web. H solutions are now all part of excel, and together we have 60 years of it, training, experience, and we all work together to bring you

3
00:00:43.740 --> 00:01:08.910
Axcel ILT: customize courses for teams from accelerate vendor, authorized certification courses from exit certified and large scale upscaling programs for your entire organization with web H solutions. And we teach a lot of different technologies. So of course, all the Microsoft official courses like fabric, but also data visualization, cloud technology, data, engineering devops and a lot more.

4
00:01:10.721 --> 00:01:24.019
Axcel ILT: But today we're going to be talking about Microsoft fabric with Fahim. Fahim is a Microsoft certified trainer, and he's got a lot of deep expertise in all things business, intelligence, machine learning.

5
00:01:24.080 --> 00:01:29.340
Axcel ILT: azure spark power, bi, copilot genai. And of course, Microsoft fabric.

6
00:01:29.901 --> 00:01:40.340
Axcel ILT: Fun. Fact, he's been a full stack developer since the crazy 1990 S. And holds numerous certifications, including azure solutions, architect and azure data engineer.

7
00:01:40.760 --> 00:01:49.890
Axcel ILT: We're also very lucky that he teaches our Microsoft fabric courses, and he's got 25 years of teaching experience. And he teaches this any standard

8
00:01:49.960 --> 00:01:55.659
Axcel ILT: Microsoft official courseware and also builds custom training courses for our clients

9
00:01:56.050 --> 00:02:19.019
Axcel ILT: and speaking of training. If you are looking for Microsoft fabric training after this webinar, so this is like a 1 h snapshot but a good next step would be microsoft official course. Dp, 600 Microsoft fabric analytics. Engineer. It's live instructor led and hands on. We've got virtual courses. If you're just one or 2 people needing to take a course

10
00:02:19.430 --> 00:02:41.229
Axcel ILT: that'll be running July 15th through 18.th That's the next one we have. But we can also create customized courses for your team or organization, and those can be online or at your site. So with that, I will let the team take it away, and I will stop share. Oh, and I'll put the Urls of the Microsoft fabrics courses into the chat.

11
00:02:41.250 --> 00:02:43.570
Axcel ILT: So let me go ahead and stop my share.

12
00:02:47.600 --> 00:02:51.819
Faheem Javed: Thank you very much. And and greetings everyone

13
00:02:52.100 --> 00:02:56.679
Faheem Javed: so hopefully. You can see my screen, if yes, please give me a thumbs up.

14
00:02:56.820 --> 00:02:57.560
Faheem Javed: I'm giving.

15
00:02:57.560 --> 00:02:58.729
Axcel ILT: Virtual thumbs up.

16
00:02:58.990 --> 00:03:16.559
Faheem Javed: Okay? Awesome. Okay? So let's get started with the agenda. So overall the the purpose of this brief webinar is to review or cover what exactly fabric is, and how it can help you or your organization to accomplish your goals.

17
00:03:16.977 --> 00:03:37.570
Faheem Javed: So, keeping this in mind. Let me quickly discuss the agenda first.st So overall it's about just a end to end overview of fabric. So what exactly fabric is, and I'll talk about what used to be the case before fabric came out. And what's the new picture of the hold since fabric has come out.

18
00:03:37.570 --> 00:03:50.190
Faheem Javed: Then I'll talk about. What if you're in a different role like you're a data engineer, you're a data modeler, slash data warehouse designer. You are a data scientist or data analyst.

19
00:03:50.200 --> 00:04:05.689
Faheem Javed: or if you are a decision maker, then how you can benefit from this fabric technology. And then, last, but not the least. I'll also go over some additional miscellaneous bits and pieces as well that are offered by this fabric technology.

20
00:04:05.840 --> 00:04:09.049
Faheem Javed: So keeping this in mind, let's get started with the

21
00:04:09.290 --> 00:04:14.570
Faheem Javed: overview of fabric. So what exactly fabric is and why we should hear for it.

22
00:04:15.405 --> 00:04:19.409
Faheem Javed: So fabric wise. It's basically an end to end

23
00:04:19.720 --> 00:04:25.109
Faheem Javed: data analytics platform end to end. Everything can be done

24
00:04:25.410 --> 00:04:49.970
Faheem Javed: right from getting your raw data to cleaning it up as a data engineer, you'll want to clean it up, you'll want to refine it, ship it up, model it. And then, after that, as a data scientist, you'll want to implement some machine learning AI sort of stuff and then not present fancy dashboards, reports to your executives and whatnot. So overall, it's an end to end data analytics platform.

25
00:04:49.990 --> 00:04:52.510
Faheem Javed: So you want to go from data

26
00:04:53.000 --> 00:05:05.969
Faheem Javed: to having satisfied customers, or even, as a let's say, a business decision maker. You want to be satisfied with what we have available to us. You want to be able to make decisions basically.

27
00:05:06.160 --> 00:05:08.670
Faheem Javed: So overall, it's a unified

28
00:05:08.710 --> 00:05:19.199
Faheem Javed: integrate system where we have lots of options available to integrate different data sources and whatnot. So it supports unified integration.

29
00:05:19.360 --> 00:05:35.120
Faheem Javed: It also lets you orchestrate. Your data flows like you can create pipelines, and you can also configure security like who should have access to what at the enterprise level. So there's some governance sort of aspect available as well.

30
00:05:35.270 --> 00:05:40.059
Faheem Javed: So overall a fabric is this end-to-end data analytics platform.

31
00:05:40.120 --> 00:06:01.359
Faheem Javed: And like, I briefly mentioned previously, you can benefit from this technology regardless of the role you are in. So as a data engineer, you can create your data pipelines. As a data scientist, you can use machine learning. You can use eda exploratory data and analytics if you want.

32
00:06:01.697 --> 00:06:14.189
Faheem Javed: You cannot utilize deep learning AI sort of stuff as well if needed. And even if you are just a report designer. You want to create reports, dashboards. Even then you can benefit from it.

33
00:06:14.625 --> 00:06:23.979
Faheem Javed: So data, architects, data modelers even executives managers, decision makers end users. They can benefit from this technology as well.

34
00:06:24.940 --> 00:06:29.649
Faheem Javed: So you can see that it's an end to end data analytics platform.

35
00:06:30.300 --> 00:06:35.480
Faheem Javed: Now, what used to be the case before fabric came out.

36
00:06:36.090 --> 00:06:38.270
Faheem Javed: So before fabric.

37
00:06:38.480 --> 00:06:58.680
Faheem Javed: Yes, we still had lots of technologies, and they sort of gave you this vague idea that yes, you can integrate them and create the end-to-end data, analytics, solutions. But fabric has made this more popular. It's it has made things easier as well.

38
00:06:58.810 --> 00:07:23.525
Faheem Javed: So before fabric came out like, I'll not talk about what used to the case. Let's say, in the 19 nineties, or even early 2,000 like, we used to have business intelligence development studio. We had integration services reporting services. We had the analysis services and whatnot. No, I'll not go that back or that much back so overall.

39
00:07:24.100 --> 00:07:26.960
Faheem Javed: even in the recent past we have had

40
00:07:27.110 --> 00:07:28.900
Faheem Javed: power bi

41
00:07:29.600 --> 00:07:36.599
Faheem Javed: service. So how many people use power bi service, please. Raise your hand or leave a message in the chat window.

42
00:07:38.640 --> 00:07:47.059
Faheem Javed: Awesome. Appreciate it. So quite a few people already have power. Be a service. So likewise we have something called azure synapse.

43
00:07:47.190 --> 00:07:50.550
Faheem Javed: Azure synapse is the big data

44
00:07:51.020 --> 00:08:01.429
Faheem Javed: analytics solution available on azure. So it has been available for quite a while now for several years, and we also have azure data storage adls as well.

45
00:08:01.940 --> 00:08:09.320
Faheem Javed: But the issue with these technologies has been, they have been somewhat technical to integrate.

46
00:08:09.530 --> 00:08:16.719
Faheem Javed: So when you want to implement data analytics at the Enterprise level, you have to integrate these things

47
00:08:16.910 --> 00:08:31.099
Faheem Javed: so overall, you end up using this platform as a service or pass sort of offering available as part of Microsoft azure and whatnot. And you have to manually integrate all of these services.

48
00:08:31.450 --> 00:08:37.829
Faheem Javed: So you need an experienced administrator who can configure these things.

49
00:08:38.260 --> 00:08:46.529
Faheem Javed: So that is where some companies usually give up on this big data solution on azure slash cloud platform.

50
00:08:46.910 --> 00:08:49.730
Faheem Javed: So that is weird. Fabric

51
00:08:50.300 --> 00:08:51.800
Faheem Javed: simplifies things.

52
00:08:51.970 --> 00:09:06.360
Faheem Javed: So Microsoft realized the fact that there will be lots of customers, clients, organizations. We will not have a very, let's say, experienced administrator who will want to manage the whole thing.

53
00:09:06.820 --> 00:09:15.429
Faheem Javed: on the daily basis. So at times you want to simplify the solution and still want to be able to implement this big data

54
00:09:15.450 --> 00:09:17.439
Faheem Javed: analytics sort of solution.

55
00:09:17.920 --> 00:09:21.149
Faheem Javed: So that is where fabric simplifies things.

56
00:09:21.290 --> 00:09:25.159
Faheem Javed: So overall, it basically provides us with the

57
00:09:25.220 --> 00:09:30.469
Faheem Javed: everything available as part of the single service called fabric.

58
00:09:30.810 --> 00:09:35.220
Faheem Javed: So now it is a software as a service. Sas.

59
00:09:35.980 --> 00:09:45.131
Faheem Javed: so overall, you just go to fabric.com, or I'll show you other options as well in just a bit. But overall you can utilize.

60
00:09:45.660 --> 00:09:55.560
Faheem Javed: this fabric subscription, and you'll have everything available to you almost out of the box, and there will be very little configuration required.

61
00:09:55.860 --> 00:10:00.220
Faheem Javed: So your main focus will be on using this out of the box

62
00:10:00.250 --> 00:10:02.800
Faheem Javed: kind of integrated environment

63
00:10:02.890 --> 00:10:17.540
Faheem Javed: that lets you do everything. Like as a data engineer, you want to create data pipelines. As a data scientist, you want to use machine learning AI sort of stuff, even generative AI or Gen AI sort of stuff and whatnot.

64
00:10:17.610 --> 00:10:22.660
Faheem Javed: So there'll be more focus on implementation and less on administration.

65
00:10:23.650 --> 00:10:31.990
Faheem Javed: and it also supports hybrid or multi-cloud integrations where you could still have your on-prem kind of infrastructure.

66
00:10:32.000 --> 00:10:39.019
Faheem Javed: and you can still integrate your on-prem setup with your on cloud fabric sort of setup.

67
00:10:39.140 --> 00:10:46.990
Faheem Javed: Or, for that matter, you could have maybe data on Aws, cloud platform, and even that can be integrated as well.

68
00:10:47.460 --> 00:10:53.659
Faheem Javed: So simply put the major difference between the older approach and the newer approach is.

69
00:10:53.700 --> 00:11:01.350
Faheem Javed: the older approach was manually integrated. You have to configure everything manually

70
00:11:03.430 --> 00:11:08.420
Faheem Javed: and no doubt with this manual integration you do get more advanced

71
00:11:08.590 --> 00:11:32.830
Faheem Javed: bells and whistles or features like you can have more advanced AI machine learning implemented, you can have, insanely huge amount of data processed as well. In a very short amount of time. So definitely, no doubt, the manual integrated solution will be more superior. But if you want ease of use

72
00:11:32.990 --> 00:11:47.900
Faheem Javed: and you don't care for just insane amount of data you have maybe a few 100 GB of data. It's not petabyte of data. In that case, fabric is more than sufficient.

73
00:11:48.200 --> 00:11:53.740
Faheem Javed: So fabric overall lets you store data, process it and present it

74
00:11:54.198 --> 00:11:57.700
Faheem Javed: to your end. Users in the form of dashboards and whatnot.

75
00:11:59.080 --> 00:12:00.812
Faheem Javed: So overall fabric.

76
00:12:02.200 --> 00:12:05.350
Faheem Javed: is assess software as a service solution.

77
00:12:05.570 --> 00:12:06.285
Faheem Javed: And

78
00:12:07.040 --> 00:12:09.839
Faheem Javed: the foundation here is this one link.

79
00:12:10.530 --> 00:12:17.070
Faheem Javed: So one lick is basically your data stored solution. So this is where you store every type of

80
00:12:17.190 --> 00:12:19.090
Faheem Javed: file you can think of.

81
00:12:19.500 --> 00:12:25.500
Faheem Javed: So if you have, let's say, excel workbooks, Csv files, Json files and whatnot.

82
00:12:25.510 --> 00:12:33.380
Faheem Javed: Even if we have SQL. Server, relational databases and whatnot. You will eventually want to store data in this one license.

83
00:12:33.500 --> 00:12:36.779
Faheem Javed: Something I'll point out in more detail in just a bit.

84
00:12:37.390 --> 00:12:41.619
Faheem Javed: So this is your data storage area, you store data here

85
00:12:42.180 --> 00:12:43.890
Faheem Javed: and the knob

86
00:12:44.285 --> 00:13:02.529
Faheem Javed: when you are in a different roles, such as data, engineer data, scientists, data, analyst and whatnot, you'll be able to use different bits and pieces. Something I'll talk about in just a bit like data factory we have other solutions or options available as well. And we even have power bi as well.

87
00:13:04.250 --> 00:13:05.922
Faheem Javed: So simply put

88
00:13:06.680 --> 00:13:07.810
Faheem Javed: fabric.

89
00:13:08.450 --> 00:13:17.109
Faheem Javed: basically this thing. So if I show you a brief demo before I discuss more slides. So if you let's say, Go to power, Bi com.

90
00:13:19.065 --> 00:13:23.620
Faheem Javed: And when you create a new workspace, so if I go to this workspaces

91
00:13:23.720 --> 00:13:29.190
Faheem Javed: and create a new workspace. Here you will notice that we have different subscriptions available.

92
00:13:29.390 --> 00:13:36.879
Faheem Javed: So, for example, we have pro. We have premium capacities there as well. We even have fabric as well.

93
00:13:37.000 --> 00:13:42.350
Faheem Javed: So if you go with this particular subscription fabric capacity, subscription.

94
00:13:42.360 --> 00:13:48.989
Faheem Javed: you'll be able to make your power Bi service portal a lot more

95
00:13:49.260 --> 00:13:50.440
Faheem Javed: powerful.

96
00:13:50.550 --> 00:13:53.689
Faheem Javed: So now you will have different sections available to you.

97
00:13:53.940 --> 00:14:00.109
Faheem Javed: So are you a data engineer? Do you want to engineer your data by creating your data pipelines.

98
00:14:00.200 --> 00:14:05.600
Faheem Javed: Are you a data scientist who wants to use maybe Jupiter notebooks

99
00:14:06.032 --> 00:14:17.580
Faheem Javed: write code in python and implement machine learning? Gen. AI, or deep learning sort of stuff and whatnot. So you can use different views.

100
00:14:18.000 --> 00:14:23.850
Faheem Javed: So we have different additional capabilities available when we have this fabric subscription.

101
00:14:24.180 --> 00:14:34.150
Faheem Javed: and I'll show you some more Demos in just a bit. But overall know that when you have the fabric subscription available you have everything end to end

102
00:14:34.440 --> 00:14:38.150
Faheem Javed: data, engineering, slash data, analytics, options available to you.

103
00:14:39.350 --> 00:14:44.260
Faheem Javed: So overall. If you are, let's say a data engineer

104
00:14:45.300 --> 00:14:52.649
Faheem Javed: fabric lets you utilize different technologies like you can use SQL, queries

105
00:14:53.050 --> 00:14:58.169
Faheem Javed: to query your data to design your data structure, your data warehouses and whatnot.

106
00:14:58.420 --> 00:15:11.859
Faheem Javed: Or you can say that. No, I'm a python based. Developer, I know python programming. I'm a data scientist. I'm a data engineer. I want to create my pipelines by using python.

107
00:15:12.040 --> 00:15:13.950
Faheem Javed: So you can use spark

108
00:15:14.250 --> 00:15:18.180
Faheem Javed: or price park option as part of this fabric technology.

109
00:15:19.440 --> 00:15:26.270
Faheem Javed: So here, if I just briefly point it out. For now, if I switch to. Let's say, maybe this demo.

110
00:15:26.420 --> 00:15:32.390
Faheem Javed: So here you can see that. I'm using fabric, and I'm able to write code in python.

111
00:15:32.830 --> 00:15:37.720
Faheem Javed: I can read files that are uploaded to my one link.

112
00:15:37.980 --> 00:15:47.299
Faheem Javed: which is the storage area available as part of fabric, so you cannot see the output as well. You can even query your other file formats.

113
00:15:47.340 --> 00:15:51.569
Faheem Javed: So, for example, here you can even have, let's say, excel workbooks as well.

114
00:15:51.660 --> 00:16:03.319
Faheem Javed: So here I already have. Let's say this sample Excel workbook. I'm reading the data. And I'm able to show outputs here in the form of tables, charts, and whatnot.

115
00:16:03.770 --> 00:16:08.419
Faheem Javed: So, in short, as a data scientist as a data engineer, we can use python

116
00:16:09.750 --> 00:16:13.352
Faheem Javed: by using this part pool which is available as part of

117
00:16:14.100 --> 00:16:15.620
Faheem Javed: fabric technology.

118
00:16:15.850 --> 00:16:37.539
Faheem Javed: And again, no special setup is required. You just need to get the subscription. And now Microsoft will take care of everything for you, like the setup of the infrastructure integration of those technology that is mostly done by Microsoft. And you just focus on writing your code in Python and SQL. And just get done with it.

119
00:16:38.430 --> 00:16:53.299
Faheem Javed: So later on, I'll point point out some SQL. Sort of stuff as well, but overall know that as an SQL. Developer, or even as a python data scientist data engineer, you can benefit from this fabric technology.

120
00:16:53.880 --> 00:17:04.689
Faheem Javed: If you are in some other role, you can use some other technologies like, if you are a data analyst. You just want to design reports, create dashboards.

121
00:17:04.780 --> 00:17:07.760
Faheem Javed: We have power bi service available

122
00:17:07.849 --> 00:17:10.159
Faheem Javed: as part of the same fabric technology.

123
00:17:10.540 --> 00:17:15.359
Faheem Javed: So those who have already used power Bi before know that it's available

124
00:17:15.970 --> 00:17:18.669
Faheem Javed: here in fabric, along with

125
00:17:19.089 --> 00:17:26.799
Faheem Javed: the rest of the components of this ecosystem that lets you store data as part of one lit

126
00:17:28.119 --> 00:17:37.850
Faheem Javed: where you can store data in various file formats. Like, usually advanced data. Engineers prefer using Delta lake

127
00:17:37.860 --> 00:17:44.200
Faheem Javed: or part K file formats. Otherwise you can just use Csp, excel word books if you want.

128
00:17:44.460 --> 00:17:50.990
Faheem Javed: and these are the different technologies or sort of different

129
00:17:51.130 --> 00:18:07.710
Faheem Javed: services, they let us utilize power bi. We can utilize a python or pice park as part of the spark cluster. We can use a T. SQL. Or transact SQL. As well and whatnot, and then various roles can benefit from these.

130
00:18:08.350 --> 00:18:09.600
Faheem Javed: So you just

131
00:18:09.920 --> 00:18:19.959
Faheem Javed: subscribe to it. Get a subscription, and you just open up the right area where you're supposed to do some data modeling or write some piece of code and whatnot.

132
00:18:21.420 --> 00:18:25.520
Faheem Javed: So this is a typical end-to-end kind of architecture.

133
00:18:26.340 --> 00:18:32.659
Faheem Javed: where you will end up having some data entry system like you could have power apps

134
00:18:33.110 --> 00:18:45.959
Faheem Javed: or any other front end application can be there. You can have rest Apis, and they can write data to let's say your transactional databases like SQL. Article and whatnot.

135
00:18:46.140 --> 00:18:48.319
Faheem Javed: And then, rather than quoting

136
00:18:48.540 --> 00:18:51.979
Faheem Javed: this SQL. Server database right away.

137
00:18:52.090 --> 00:18:55.119
Faheem Javed: So rather than using power bi desktop

138
00:18:55.210 --> 00:18:58.069
Faheem Javed: and connecting to this database right away.

139
00:18:58.290 --> 00:19:04.449
Faheem Javed: we actually end up creating this enterprise setup like this.

140
00:19:05.710 --> 00:19:09.870
Faheem Javed: because if you connect to some database directly

141
00:19:10.140 --> 00:19:14.419
Faheem Javed: you're going to cause too much overhead on the database.

142
00:19:15.120 --> 00:19:17.370
Faheem Javed: So this database is already busy.

143
00:19:17.800 --> 00:19:21.819
Faheem Javed: You're adding updating deleting records non-stop here

144
00:19:22.360 --> 00:19:33.190
Faheem Javed: on the daily basis, hourly basis every minute. You could have new transactions. So now, if you tried to create reports against that database, you'll slow down the performance.

145
00:19:34.020 --> 00:19:47.179
Faheem Javed: And also if you have lots of data entry systems, you have lots of data sources like Csv excel word books. So maybe we have 300 Excel workbooks. We have 400 Csv points.

146
00:19:47.230 --> 00:19:56.919
Faheem Javed: So rather than making life miserable for these data analysts so and ask them to connect to every data source one at a time.

147
00:19:56.960 --> 00:20:02.150
Faheem Javed: We basically engineered this enterprise setup like this.

148
00:20:02.680 --> 00:20:06.639
Faheem Javed: So we basically end up taking data and

149
00:20:07.190 --> 00:20:10.170
Faheem Javed: we end up storing it in this one leg.

150
00:20:10.450 --> 00:20:22.820
Faheem Javed: So even if we have some database server, we usually end up taking data that we have in database tables, and we end up designing pipelines or data flows, and they will extract the data

151
00:20:23.120 --> 00:20:28.876
Faheem Javed: potentially, transform it, ship it up, modify it, clean it up, and then dump it

152
00:20:29.290 --> 00:20:31.279
Faheem Javed: in this one leg storage.

153
00:20:32.250 --> 00:20:36.025
Faheem Javed: and then next you can do some data modeling. You can

154
00:20:36.380 --> 00:20:38.430
Faheem Javed: interconnect tables.

155
00:20:38.520 --> 00:20:44.979
Faheem Javed: add missing columns that do not already exist. You can add them remove something if needed.

156
00:20:45.010 --> 00:20:47.709
Faheem Javed: Do data modeling here, basically.

157
00:20:47.920 --> 00:20:59.699
Faheem Javed: you can also create groups, you can create bins. So grouping and binning can be done. You can create hierarchies, and whatnot. So you can basically model your data here.

158
00:21:00.380 --> 00:21:03.039
Faheem Javed: And then, once we have this modeled

159
00:21:03.190 --> 00:21:04.810
Faheem Javed: data warehouse

160
00:21:05.364 --> 00:21:09.210
Faheem Javed: which you can also think of as Data Lake House.

161
00:21:09.350 --> 00:21:13.460
Faheem Javed: because the underlying data storage is, there's 1 link.

162
00:21:13.510 --> 00:21:17.940
Faheem Javed: So we usually call this a data lake house as well.

163
00:21:18.000 --> 00:21:20.090
Faheem Javed: Or you can just call it data warehouse.

164
00:21:20.120 --> 00:21:24.300
Faheem Javed: So simply put this database as consolidating

165
00:21:24.330 --> 00:21:27.829
Faheem Javed: data from several different data sources.

166
00:21:28.550 --> 00:21:31.199
Faheem Javed: And this is going to act as a

167
00:21:31.730 --> 00:21:33.830
Faheem Javed: single source of truth

168
00:21:33.930 --> 00:21:36.419
Faheem Javed: for your entire organization.

169
00:21:37.770 --> 00:21:41.169
Faheem Javed: So now you're asking all data analysts

170
00:21:41.290 --> 00:21:47.109
Faheem Javed: to just be aware of this single point of contact. So just this database hubs

171
00:21:47.880 --> 00:21:51.750
Faheem Javed: without worrying about where exactly the data came from.

172
00:21:51.940 --> 00:22:05.609
Faheem Javed: So data engineers, they have already created these pipelines to design this Lakehouse slash warehouse. And now, data analysts they just have to connect to this single source of truth which is insanely fast.

173
00:22:06.160 --> 00:22:10.129
Faheem Javed: So reports dashboards. They'll be very, very fast

174
00:22:10.380 --> 00:22:13.270
Faheem Javed: because of the underlying infrastructure

175
00:22:13.390 --> 00:22:20.799
Faheem Javed: which is composed of multiple nodes. They are doing massive parallel processing for us here.

176
00:22:20.830 --> 00:22:25.400
Faheem Javed: So overall, you have this enterprise setup where data is centralized

177
00:22:25.510 --> 00:22:31.559
Faheem Javed: in this our data warehouse, Slash Lake House, and it it will give you top notch performance.

178
00:22:31.690 --> 00:22:38.930
Faheem Javed: So now you can design your reports and then up, upload them, or publish them to this power bi service.

179
00:22:40.070 --> 00:22:43.700
Faheem Javed: And all of this stuff is fabric.

180
00:22:44.200 --> 00:22:46.349
Faheem Javed: It's your data storage.

181
00:22:47.212 --> 00:22:48.137
Faheem Javed: Sort of

182
00:22:48.620 --> 00:22:53.740
Faheem Javed: area. It also lets you create these pipelines to orchestrate data.

183
00:22:53.770 --> 00:22:59.160
Faheem Javed: It also lets you design your delivery houses, enterprise delivery houses.

184
00:22:59.210 --> 00:23:04.150
Faheem Javed: It lets you create reports. And it even lets data scientists

185
00:23:04.340 --> 00:23:09.650
Faheem Javed: implement machine learning AI sort of stuff that I'll briefly point out in just a bit.

186
00:23:10.040 --> 00:23:12.810
Faheem Javed: But in short, fabric is this thing.

187
00:23:13.590 --> 00:23:23.370
Faheem Javed: and you can get up and running with fabric the moment you get a subscription you have this infrastructure available to you.

188
00:23:24.130 --> 00:23:28.993
Faheem Javed: Otherwise, without fabric. Admins, they typically have to spend

189
00:23:29.992 --> 00:23:36.839
Faheem Javed: a day or 2 at least, to set up the infrastructure, then they have to regularly maintain it as well.

190
00:23:37.010 --> 00:23:51.310
Faheem Javed: whereas here you get up, and running with this fabrics setup in a very fast manner. And now maintenance wise. There's very little stuff you have to do here, so you mostly now just have to engineer your data and create.

191
00:27:17.240 --> 00:27:20.170
Faheem Javed: Hello! Testing testing can you folks hear me?

192
00:27:20.920 --> 00:27:22.009
Faheem Javed: I think I'll.

193
00:27:23.350 --> 00:27:25.009
Axcel ILT: Yes, yes, you're back.

194
00:27:25.370 --> 00:27:32.299
Faheem Javed: So yeah, that's I think I got jinxed so my machine froze up as well, and I had to restart it. Sorry about that.

195
00:27:32.310 --> 00:27:36.310
Faheem Javed: Okay, and just to double check. Am I sharing my screen.

196
00:27:37.970 --> 00:27:38.630
Axcel ILT: Yes.

197
00:27:38.890 --> 00:27:48.170
Faheem Javed: Okay? So where were we? So basically, almost any role can benefit from this fabric technology, like, as a data engineer.

198
00:27:48.480 --> 00:28:01.300
Faheem Javed: If you want to. Let's say, take data from SQL. Server and other file formats and store it in that one lake technology, you can choose to write code in Tc cool. If you want

199
00:28:01.490 --> 00:28:10.969
Faheem Javed: not that we will cover any of this stuff practically in this very brief webinar. But in the detailed courses. We do talk about all of this stuff practically.

200
00:28:11.050 --> 00:28:16.260
Faheem Javed: But for now know that you can design your data warehouse

201
00:28:16.630 --> 00:28:24.619
Faheem Javed: by consolidating, consolidating data from different data sources, either by using Tsql syntax.

202
00:28:24.880 --> 00:28:28.700
Faheem Javed: Or if you say that no, I don't know Tsql.

203
00:28:29.190 --> 00:28:37.210
Faheem Javed: T's equal, I just want to use some visual graphical technique. I want to drag and drop. I want to create my data pipeline

204
00:28:37.450 --> 00:28:44.899
Faheem Javed: in a visual manner. Then in that case you can use this data factory option azure data factory.

205
00:28:45.200 --> 00:28:52.779
Faheem Javed: So it's a graphical way of creating your pipeline. Where you'll say I want to get data from some SQL server

206
00:28:52.890 --> 00:29:05.119
Faheem Javed: or from Csp or excel. I want to transform it. Add some column, remove some column, filter out some data, and then I want to write it to my data warehouse or my lake house.

207
00:29:05.270 --> 00:29:10.959
Faheem Javed: So you can basically design your data pipelines in a graphical manner as well if needed.

208
00:29:11.260 --> 00:29:17.720
Faheem Javed: and it still lets you execute some code as well like T. Sequel code, python code if needed.

209
00:29:17.750 --> 00:29:21.470
Faheem Javed: So you still have lots of flexibility available here.

210
00:29:21.940 --> 00:29:25.909
Faheem Javed: and if you say that no, I want to use

211
00:29:26.310 --> 00:29:28.220
Faheem Javed: powered BI

212
00:29:29.420 --> 00:29:30.490
Faheem Javed: styled

213
00:29:30.550 --> 00:29:50.400
Faheem Javed: option to create my let's say, data warehouse. I want to take data from my data sources like excel Csv. SQL. Server and whatnot, and I want to modify it. Shape it up, and I want to dump it to my data warehouse. You can use this option, as well called data, flows

214
00:29:51.113 --> 00:29:55.850
Faheem Javed: question. How many people are familiar with power? Query, editor.

215
00:29:56.390 --> 00:30:00.720
Faheem Javed: has anyone utilized power? Query, editor in power, bi

216
00:30:01.030 --> 00:30:03.019
Faheem Javed: desktop, or something like that.

217
00:30:07.860 --> 00:30:15.210
Faheem Javed: Yes, awesome. So those who have used this option know that you can use the exact same ui, the exact same look and feel.

218
00:30:15.786 --> 00:30:18.320
Faheem Javed: To create your data pipelines.

219
00:30:20.150 --> 00:30:36.640
Faheem Javed: And even if you're using, let's say tableau, also a tableau. Wise? No, when it comes to creating your pipeline. No, you cannot use tableau at this point. You can use tableau to design your reports and dashboards afterward. But no to create your data pipelines.

220
00:30:36.790 --> 00:30:43.089
Faheem Javed: The 3rd option here is to use this power query, editor, sort of look and feel

221
00:30:43.160 --> 00:30:50.100
Faheem Javed: so overall data. Pipelines can be designed in several ways. Case equal

222
00:30:50.569 --> 00:30:57.870
Faheem Javed: graphical technique power query, editor based technique. Or if you say that I'm a python developer.

223
00:30:58.050 --> 00:31:05.800
Faheem Javed: I'm a data scientist. I'm a data engineer who knows python. Then, in that case you cannot use Jupiter notebooks

224
00:31:05.870 --> 00:31:08.490
Faheem Javed: and write code in python. Of

225
00:31:08.560 --> 00:31:27.630
Faheem Javed: that basically ends up utilizing the spark cluster. So overall, we call it pi spark, and you'll be able to read any file, any data source of your choice, do any sort of activities such as any transformation can be performed here, and you can dump it to any destination of your choice.

226
00:31:27.760 --> 00:31:32.800
Faheem Javed: So even as a python programmer, you can create your data pipelines.

227
00:31:33.608 --> 00:31:37.790
Faheem Javed: Is anyone a python programmer here, or a data scientist?

228
00:31:41.330 --> 00:31:45.170
Faheem Javed: So those who use python, please raise your hand or leave a message in the chat window.

229
00:31:50.080 --> 00:32:01.079
Faheem Javed: and while we are talking about this python portion, let me just briefly point it out here. So overall, when you create a workspace here that is making use of

230
00:32:01.100 --> 00:32:03.820
Faheem Javed: fabric, the fabric subscription.

231
00:32:03.870 --> 00:32:09.700
Faheem Javed: You'll be able to create your data links from here, or let me just switch to this one. For now.

232
00:32:11.114 --> 00:32:14.665
Faheem Javed: This one and let me open up. This

233
00:32:16.440 --> 00:32:19.480
Faheem Javed: Lake House, or I think I already have this one open here.

234
00:32:20.100 --> 00:32:21.930
Faheem Javed: perhaps this one.

235
00:32:22.850 --> 00:32:37.720
Faheem Javed: Okay. So when you. Create a fabric based subscription, or you have the fabric subscription available to you. You'll be able to create new Jupiter notebooks here, and you'll be able to utilize Price Park, or even other languages if needed

236
00:32:38.150 --> 00:32:48.589
Faheem Javed: so definitely. If you're familiar with python pandas you use Mac flot lip. You use a seaborne or the pytorch.

237
00:32:48.750 --> 00:32:58.619
Faheem Javed: So there, these are different libraries we use for machine learning, data, analytics and whatnot. You are able to use all of that stuff here. You can write the code in python here.

238
00:33:00.500 --> 00:33:15.799
Faheem Javed: and you don't need to set up anything in particular. You just subscribe. Get the subscription, basically. So just get the fabric subscription, and you already have this integrated spark pool available where you can start creating your Jupiter notebooks

239
00:33:16.450 --> 00:33:25.629
Faheem Javed: so simply port. You can write code in python so that you can design your pipelines to create your data warehouse and whatnot

240
00:33:27.858 --> 00:33:34.671
Faheem Javed: and if you want to use sort of a more visual way of setting up your let's see,

241
00:33:35.360 --> 00:33:44.770
Faheem Javed: data model like, you want to create a relationship between tables and whatnot. You can design your data warehouse in a more graphical manner as well.

242
00:33:44.880 --> 00:33:55.819
Faheem Javed: So you'll be able to see what one lake storage based. Files are available to you, and then you'll be able to convert them to tables

243
00:33:55.950 --> 00:34:06.349
Faheem Javed: very much like database tables, and you'll be able to interconnect these tables by using one to 1, one, to many, many to one, many to many sort of relationships.

244
00:34:06.370 --> 00:34:12.959
Faheem Javed: So you can design your Databath houses in sort of a more graphical manner as well.

245
00:34:13.500 --> 00:34:36.310
Faheem Javed: So you can see that if you are a beginner or a non technical user, you just want to do things graphically. You can do so if you are familiar with T. Sequel. You can do that as well. If you're familiar with the python, you can use that as well. So it offers different technologies or different features to different roles.

246
00:34:36.400 --> 00:34:41.010
Faheem Javed: so you can design your models in an easy manner as well.

247
00:34:42.110 --> 00:34:46.469
Faheem Javed: And likewise, if you are, let's say utilizing

248
00:34:46.850 --> 00:34:49.330
Faheem Javed: machine learning sort of concepts

249
00:34:50.973 --> 00:34:52.659
Faheem Javed: pattern recognition.

250
00:34:54.510 --> 00:34:56.750
Faheem Javed: such as data classification.

251
00:34:57.080 --> 00:34:59.809
Faheem Javed: So you want to find out whether

252
00:34:59.940 --> 00:35:07.389
Faheem Javed: some credit card transaction. It's a fraudulent activity, or it's a legitimate kind of transaction. You want to classify it.

253
00:35:07.460 --> 00:35:09.940
Faheem Javed: or you want to predict something.

254
00:35:10.770 --> 00:35:21.599
Faheem Javed: You want to find out what could be the next potential sale. So if we have certain sales data available to us, what could be your next potential sale available?

255
00:35:21.820 --> 00:35:32.379
Faheem Javed: You can also do some clustering as well. So, in short, they are different algorithms, machine learning sort of algorithms. And they can all be utilized to you.

256
00:35:32.800 --> 00:35:36.259
Faheem Javed: So simply put as a data scientist, you can

257
00:35:36.850 --> 00:35:49.870
Faheem Javed: perform your data. Scientist, kind of operations here like you can prepare your data by fine tuning it, which is also known as data wrangling. You can train your model.

258
00:35:50.030 --> 00:35:59.109
Faheem Javed: which is a bunch of algorithms. You can do hyper parameter tuning as well. You can score your models and get the predictions.

259
00:35:59.420 --> 00:36:01.929
Faheem Javed: And actually, none of this stuff is new

260
00:36:01.960 --> 00:36:06.169
Faheem Javed: like, although these days we mostly use python.

261
00:36:06.240 --> 00:36:10.320
Faheem Javed: but it has been available for a very, very long time.

262
00:36:10.790 --> 00:36:14.259
Faheem Javed: So machine learning has been available for a very, very long time.

263
00:36:14.780 --> 00:36:23.619
Faheem Javed: Like, personally, I've been using machine learning even as part of business Intelligence Development studio 2,005.

264
00:36:24.120 --> 00:36:27.989
Faheem Javed: So even then, we had something called data mining.

265
00:36:28.000 --> 00:36:43.020
Faheem Javed: where we used to create a mining structures, we used to specify different algorithms like neural network association clustering and whatnot and what exactly you want to predict. We used to specify that

266
00:36:43.040 --> 00:36:56.829
Faheem Javed: we were able to use Dmx data, mining extension syntax to specify what algorithm or model we want to use such as decision trees and whatnot. And we were able to predict data.

267
00:36:56.970 --> 00:37:04.499
Faheem Javed: So even back in the day, we had a machine learning available, and these days we mostly do it in python.

268
00:37:04.560 --> 00:37:10.189
Faheem Javed: So python wise. Here I have this simple machine learning sort of scenario.

269
00:37:10.250 --> 00:37:18.289
Faheem Javed: where it will analyze a Csv file and perform a semantic

270
00:37:18.690 --> 00:37:22.379
Faheem Javed: or it will basically look at. The

271
00:37:22.855 --> 00:37:30.179
Faheem Javed: lines of code that I have there. And it will basically check what is the sentiment sentiment analysis? Basically.

272
00:37:30.240 --> 00:37:33.080
Faheem Javed: So we are reading a text file in this case.

273
00:37:33.150 --> 00:37:39.039
Faheem Javed: and not that we have enough time as part of this webinar. But I'll just briefly point out

274
00:37:39.660 --> 00:37:43.500
Faheem Javed: at this very simple data set here where we have

275
00:37:43.610 --> 00:37:45.869
Faheem Javed: some product

276
00:37:46.200 --> 00:37:48.300
Faheem Javed: specific comments

277
00:37:48.440 --> 00:37:50.219
Faheem Javed: along with the sentiment.

278
00:37:51.260 --> 00:37:56.959
Faheem Javed: So if let's say we have. I loved this product. It's amazing. It's a positive sentiment, sentiment.

279
00:37:57.070 --> 00:38:03.089
Faheem Javed: If something is terrible, it's a negative sentiment. So we have positive negative sentiment sealed.

280
00:38:03.290 --> 00:38:14.699
Faheem Javed: So if you want to do sentiment analysis, we can use machine learning for this purpose, and we can predict what could be the next potential comment posted by some user.

281
00:38:14.750 --> 00:38:23.670
Faheem Javed: So here we are basically training our model, we are scoring it. And we are coming up with some prediction or the accuracy for that model as well.

282
00:38:23.880 --> 00:38:36.970
Faheem Javed: And back in the day we had the same stuff. Available as product, business intelligence, development studio slash. Ss, where we were able to do data mining use different machine learning algorithms and what? Whatnot.

283
00:38:37.210 --> 00:38:47.310
Faheem Javed: So you can basically do this thing in fabric by writing code in python without focusing on setting up things.

284
00:38:47.530 --> 00:38:51.279
Faheem Javed: So just start creating your Jupiter notebooks.

285
00:38:51.300 --> 00:38:58.380
Faheem Javed: Write your code and whatever your data set is available to you like your one league

286
00:38:58.690 --> 00:39:03.850
Faheem Javed: or database house sort of stuff. You basically just analyze that.

287
00:39:04.440 --> 00:39:06.640
Faheem Javed: Make predictions and whatnot.

288
00:39:07.570 --> 00:39:12.169
Faheem Javed: So simply put all of this stuff is available out of the box.

289
00:39:12.480 --> 00:39:14.160
Faheem Javed: As a data scientist.

290
00:39:14.300 --> 00:39:18.469
Faheem Javed: you're able to. Do this stuff out of the box.

291
00:39:20.350 --> 00:39:25.429
Faheem Javed: Okay? Does anyone use data science concepts. Do you usually

292
00:39:25.520 --> 00:39:27.960
Faheem Javed: utilize machine learning sort of concepts

293
00:39:28.810 --> 00:39:30.420
Faheem Javed: or have any interest in it?

294
00:39:31.390 --> 00:39:38.940
Faheem Javed: You can now post that in the chart window. Yes. So those who want to do this stuff fabric already has this stuff.

295
00:39:39.220 --> 00:39:42.310
Faheem Javed: Those who don't care for it skip it.

296
00:39:42.530 --> 00:39:46.810
Faheem Javed: You can still benefit from other areas. Other roles

297
00:39:46.890 --> 00:39:50.180
Faheem Javed: that are there in this fabric technology.

298
00:39:50.970 --> 00:39:55.030
Faheem Javed: And those who just want to create reports.

299
00:39:56.573 --> 00:39:57.919
Faheem Javed: and dashboards.

300
00:39:58.190 --> 00:40:00.300
Faheem Javed: You can use power bi

301
00:40:00.420 --> 00:40:02.440
Faheem Javed: and connect to that

302
00:40:02.460 --> 00:40:04.169
Faheem Javed: single source of truth

303
00:40:04.410 --> 00:40:07.810
Faheem Javed: to one lake, basically, or your database house

304
00:40:07.930 --> 00:40:11.159
Faheem Javed: slash late house that I talked about previously.

305
00:40:11.260 --> 00:40:16.259
Faheem Javed: So fabric already has the database that is your single source of truth.

306
00:40:16.950 --> 00:40:23.620
Faheem Javed: You don't need to focus on 300 Csv files, 400 Excel workbooks.

307
00:40:23.630 --> 00:40:27.399
Faheem Javed: 30 SQL. Server article databases anymore.

308
00:40:27.510 --> 00:40:30.899
Faheem Javed: You just focus on your single source of truth

309
00:40:31.140 --> 00:40:32.380
Faheem Javed: data warehouse.

310
00:40:32.630 --> 00:40:34.460
Faheem Javed: which is super fast.

311
00:40:35.250 --> 00:40:38.189
Faheem Javed: When it comes to analyzing data, it's very, very fast.

312
00:40:38.700 --> 00:40:44.159
Faheem Javed: So you're basically able to create reports by using power bi desktop

313
00:40:44.180 --> 00:40:48.450
Faheem Javed: and fabric already has power bi service.

314
00:40:49.140 --> 00:40:52.900
Faheem Javed: So you can publish your reports to the same power bi service

315
00:40:52.970 --> 00:40:56.480
Faheem Javed: which is available as part of fabric.

316
00:40:56.980 --> 00:41:04.600
Faheem Javed: So, for example, even here, if I switch back to this fabric. So here we can see that we have

317
00:41:05.490 --> 00:41:09.059
Faheem Javed: the ability to deploy or publish reports as well.

318
00:41:09.070 --> 00:41:28.659
Faheem Javed: So for now I'll just go to this particular Workspace, and you can see that I have a bunch of reports already published, and they are connecting to the underlying database, Slash Lake House, and you are able to organize your report in the form of pages. Ignore the page names for now. So we have a bunch of pages here.

319
00:41:30.150 --> 00:41:37.539
Faheem Javed: So you're able to create reports, and they are able to use that single source of truth, which is very fast.

320
00:41:39.400 --> 00:41:43.560
Faheem Javed: and you can also create dashboards down the road

321
00:41:43.960 --> 00:41:47.940
Faheem Javed: where each dashboard is a web page.

322
00:41:48.210 --> 00:41:50.300
Faheem Javed: and it's able to get

323
00:41:50.780 --> 00:41:53.370
Faheem Javed: visuals from different reports.

324
00:41:54.750 --> 00:42:02.719
Faheem Javed: So, for example, your executives, managers, decision makers, they will not have the time to go through every detailed report.

325
00:42:03.010 --> 00:42:06.419
Faheem Javed: They will say, create more focused

326
00:42:06.910 --> 00:42:08.479
Faheem Javed: role specific

327
00:42:08.680 --> 00:42:09.720
Faheem Javed: of

328
00:42:10.870 --> 00:42:12.830
Faheem Javed: dashboards here.

329
00:42:13.450 --> 00:42:17.379
Faheem Javed: So each dashboard is a web page where you basically just

330
00:42:17.730 --> 00:42:22.200
Faheem Javed: add bits and pieces from different reports. And you say, this is my dashboard.

331
00:42:24.520 --> 00:42:25.869
Faheem Javed: So here to

332
00:42:26.010 --> 00:42:30.179
Faheem Javed: if I just briefly show you this thing, note that we have

333
00:42:30.720 --> 00:42:35.029
Faheem Javed: to do this thing in detail here, in this course or this webinar.

334
00:42:35.060 --> 00:42:59.450
Faheem Javed: but in 30 full blown courses we usually cover all of the stuff practically so here I have this sample dashboard. Each visual is coming from a different report, and you're able to arrange these in the form of tiles. You can move these around, structure your dashboard, however, you want. People can even subscribe to it, so that they will get notified whenever any changes have been made to the underlying data and whatnot.

335
00:42:59.620 --> 00:43:03.580
Faheem Javed: So, and you can also navigate to your detailed report from here, if needed.

336
00:43:04.140 --> 00:43:10.629
Faheem Javed: So simply port, you can visualize data presented to your managers, decision makers and whatnot.

337
00:43:14.620 --> 00:43:23.610
Faheem Javed: so? Even if just report designing is your main focus. Fabric is still helpful. You can connect to your underlying data source and whatnot

338
00:43:24.130 --> 00:43:28.729
Faheem Javed: in more advanced cases. You can also create apps

339
00:43:29.420 --> 00:43:30.940
Faheem Javed: power bi apps.

340
00:43:32.240 --> 00:43:39.460
Faheem Javed: So power bi app we don't have the time to see that practically. But I'll just show you the screenshot, for now.

341
00:43:39.610 --> 00:43:43.879
Faheem Javed: in the full blown course, we see it practically we discuss how to design it.

342
00:43:44.840 --> 00:43:46.860
Faheem Javed: You want to package

343
00:43:46.870 --> 00:43:48.820
Faheem Javed: all of your reports.

344
00:43:49.440 --> 00:43:51.040
Faheem Javed: your dashboards.

345
00:43:51.480 --> 00:43:56.590
Faheem Javed: your Kpis. They're also known as score cards.

346
00:43:57.010 --> 00:44:01.489
Faheem Javed: So you want to package everything that is relevant to

347
00:44:01.520 --> 00:44:03.060
Faheem Javed: your audience.

348
00:44:03.880 --> 00:44:31.669
Faheem Javed: and you want to present everything in sort of an application format. So here we have something called power bi app, where each item here is some report or dashboard, or some scorecard. And now end users can look at it, your executives, your outsiders, and whatnot like customers, clients, and whatnot. They can look at this app. They can go through these reports, dashboards that are relevant to them.

349
00:44:31.760 --> 00:44:36.150
Faheem Javed: So you can basically bundle up everything, package up everything as a power bi app.

350
00:44:36.160 --> 00:44:38.539
Faheem Javed: even that is supported by fabric.

351
00:44:39.370 --> 00:44:47.760
Faheem Javed: So, in short, overall fabric. It's a huge technology. It even has some other miscellaneous bits and pieces like you can

352
00:44:48.000 --> 00:45:00.169
Faheem Javed: configure some triggers like. If data goes below some threshold amount or over some threshold amount. You want to trigger some workflow.

353
00:45:00.220 --> 00:45:05.160
Faheem Javed: such as some, maybe power automate kind of activity can be executed.

354
00:45:05.380 --> 00:45:31.919
Faheem Javed: So you can basically trigger some action. You can even receive notifications as well that something meets your expectations, or it has gone above the expectation, expected value or below the expected value and whatnot. So you can use get activated for this purpose as part of fabric. And we can even analyze real time streaming kind of data as well. Nonstop continuous dashboards.

355
00:45:32.540 --> 00:45:35.969
Faheem Javed: or a dashboard updates can be implemented as well.

356
00:45:36.830 --> 00:45:42.270
Faheem Javed: So in short, overall fabric, it's a huge technology. And there's a lot to it.

357
00:45:42.330 --> 00:45:44.980
Faheem Javed: So this is over on the end of

358
00:45:46.280 --> 00:45:51.380
Faheem Javed: end-to-end sort of data, engineering, slash data, analytics, sort of stuff.

359
00:45:51.700 --> 00:45:52.713
Faheem Javed: Okay, so

360
00:45:53.390 --> 00:45:55.360
Faheem Javed: any questions regarding this part.

361
00:45:55.640 --> 00:45:56.820
Faheem Javed: this webinar

362
00:45:58.170 --> 00:46:08.050
Faheem Javed: and yes, we'll be providing you with the slide deck as well as the video recording. And we do have detailed courses available on this technology.

363
00:46:08.200 --> 00:46:13.619
Faheem Javed: like, if you want to look at sort of the whole thing, we have DP. 600.

364
00:46:14.199 --> 00:46:39.930
Faheem Javed: Otherwise, you can also choose to pick and choose certain areas like, do you want to focus on Tsql portion? Do you want to focus on the python portion? Do you want to be a data scientist or machine learning sort of programmer? Or, let's, let's just call it data science area. So do you want to do that on a fabric? Do you want to just focus on reports? So we have various

365
00:46:40.350 --> 00:46:42.950
Faheem Javed: sort of individual courses available?

366
00:46:43.412 --> 00:46:49.749
Faheem Javed: Or if you want to use azure synapse, which is the more sort of

367
00:46:49.980 --> 00:47:09.969
Faheem Javed: advanced way of implementing fabric. But you have to administrator or administer everything manually, you will have to integrate everything manually. So in that case you will have to use a data engineering option on Microsoft azure by using synapse. So you'll end up using synapse. Basically. So for that, we have this particular course available.

368
00:47:11.136 --> 00:47:13.920
Faheem Javed: Okay, so any questions regarding

369
00:47:14.060 --> 00:47:14.665
Faheem Javed: this

370
00:47:15.720 --> 00:47:16.810
Faheem Javed: session.

371
00:47:18.840 --> 00:47:20.969
Faheem Javed: I'm going to check the chat window.

372
00:47:28.470 --> 00:47:39.629
Faheem Javed: So simply port, it's just the end to end data engineering, slash data, analytics, slash machine learning like data, scientist sort of area slash data.

373
00:47:40.063 --> 00:47:48.120
Faheem Javed: Analyst kind of area. Even our decision makers can look at the picture or the end goal as well. And whatnot.

374
00:47:48.140 --> 00:47:49.549
Faheem Javed: So it's basically this

375
00:47:50.210 --> 00:47:53.919
Faheem Javed: portion which integrates all of these important technologies

376
00:47:54.040 --> 00:48:14.830
Faheem Javed: in a very easy manner. Where you don't focus on the setup part, you instead focus on implementing it. You write code in python you use pandas. You use pytorch and various other libraries and frameworks in python. Or you can say, I just want to use T sequel, or just a graphical

377
00:48:14.980 --> 00:48:18.299
Faheem Javed: tool set to design my data, pipelines and whatnot.

378
00:48:19.690 --> 00:48:27.879
Faheem Javed: There's a question in the chart window, what are the commercial aspects like, okay, so this is basically a software as a service offering.

379
00:48:28.050 --> 00:48:30.389
Faheem Javed: So this is Sas offering.

380
00:48:30.460 --> 00:48:32.940
Faheem Javed: Whereas if you are using synaps

381
00:48:33.870 --> 00:48:41.310
Faheem Javed: and you integrated that with power Bi and azure data storage. That is the platform as a service offering.

382
00:48:41.880 --> 00:48:47.970
Faheem Javed: So this is a software as a service offering, where you just get the subscription. And now you'll be billed.

383
00:48:48.190 --> 00:48:52.429
Faheem Javed: not nonstop, but only for the actual consumption.

384
00:48:52.830 --> 00:48:58.099
Faheem Javed: So when your let's say, Jupiter Notebook containing python code is running.

385
00:48:58.800 --> 00:49:05.529
Faheem Javed: whatever the amount of resources will be utilized to process that you only pay for that actual processing.

386
00:49:05.570 --> 00:49:15.680
Faheem Javed: So you're not paying for it. Nonstop like when you're sleeping, or something like that. So it's not 24, 7 kind of cost. It's only the cost for the actual consumption.

387
00:49:15.830 --> 00:49:18.010
Faheem Javed: So that is basically software as a service.

388
00:49:18.696 --> 00:49:22.950
Faheem Javed: So there's no initial cost at all. There's no upfront fee.

389
00:49:23.090 --> 00:49:30.349
Faheem Javed: It's only built by the hour, and then you have to pay at the end of the month. So it's for the actual consumption.

390
00:49:30.620 --> 00:49:32.740
Faheem Javed: Unlike synapse.

391
00:49:32.760 --> 00:49:37.000
Faheem Javed: Where? When you set up things now, you typically have to pay for it.

392
00:49:37.447 --> 00:49:42.869
Faheem Javed: There's certain amount you have to pay for it unless you stop it. Manually.

393
00:49:43.100 --> 00:49:49.019
Faheem Javed: Same goes for azure data like storage. You keep paying for that data lake storage

394
00:49:49.030 --> 00:49:51.219
Faheem Javed: based on the data size and whatnot.

395
00:49:51.520 --> 00:49:57.859
Faheem Javed: So overall, there's no upfront fee. It's a basically pay as you go, sort of option.

396
00:50:00.260 --> 00:50:12.809
Faheem Javed: And yes, you can definitely connect to your on-prem setup as well. So for that, it supports data gateway option. So you can set up your gateways, data gateways that can be in your on-prem environment, or even on azure

397
00:50:12.820 --> 00:50:17.220
Faheem Javed: or let's say, or even on aws as well. So yes, it does support

398
00:50:17.380 --> 00:50:22.070
Faheem Javed: integration like a hybrid as well as cross cloud

399
00:50:22.330 --> 00:50:24.530
Faheem Javed: sort of setup can be implemented as well.

400
00:50:26.840 --> 00:50:47.659
Faheem Javed: So a very good question regarding the limitations and whatnot. So overall you basically have to upload everything to this one lake. So even this data warehouse, you will end up creating here. It actually stores data in this one lake. So one lake wise overall, there isn't any limitation as such.

401
00:50:47.870 --> 00:50:56.540
Faheem Javed: But the resources that will be available to you. Typically, they are okay for around, let's say, a 10 TB of data.

402
00:50:56.730 --> 00:51:02.590
Faheem Javed: So if you are close to 10 TB of data, the resources that will be provided to you by Microsoft.

403
00:51:02.620 --> 00:51:04.349
Faheem Javed: they are typically sufficient.

404
00:51:04.470 --> 00:51:06.639
Faheem Javed: But if you say that? No, we have

405
00:51:06.800 --> 00:51:14.630
Faheem Javed: even larger gigantic data sets. In that case you have to go with azure synapse. So synapse is meant for

406
00:51:14.670 --> 00:51:17.599
Faheem Javed: insane amount of processing capabilities.

407
00:51:17.660 --> 00:51:21.739
Faheem Javed: whereas fabric is meant for sort of

408
00:51:21.870 --> 00:51:28.500
Faheem Javed: small to mid-level sort of scenarios where you have around 10 TB of data.

409
00:51:28.760 --> 00:51:31.969
Faheem Javed: Anything over that. Yes, you can process it, but

410
00:51:32.080 --> 00:51:34.820
Faheem Javed: it will not be as fast as using synapse.

411
00:51:37.620 --> 00:51:41.460
Faheem Javed: so still preferred using synapse for more larger enterprise. Scenarios

412
00:51:42.220 --> 00:51:53.510
Faheem Javed: and fabric can be used for the majority of the scenarios. So when you hit some limit using synapse or sorry using fabric, then you can consider using synapse.

413
00:51:53.720 --> 00:51:56.339
Faheem Javed: Okay? So any further questions regarding this part.

414
00:52:02.950 --> 00:52:05.391
Faheem Javed: so fabric is available as service in

415
00:52:05.920 --> 00:52:24.959
Faheem Javed: or it uses azure as the overall infrastructure. But overall, you just have to go to, let's say, power bi.com, or there are other links that I'll share with with you later on. But overall, you can just basically navigate to even power bi.com and just get the subscription.

416
00:52:25.390 --> 00:52:40.269
Faheem Javed: So fabric is available as part of power bi.com and also as a sub as a separate yeah, URL, as well. So no, you don't need to go to the azure portal itself. You just need to go to the fabric portal. So it has its separate portal.

417
00:52:42.300 --> 00:52:50.133
Faheem Javed: So simply put, just go to power Bi Com, and you'll be able to see the subscription option for fabric. Okay? So.

418
00:52:52.030 --> 00:52:57.439
Faheem Javed: in that case. Thank you very much. It was a pleasure to be here with you for this session.

419
00:52:58.870 --> 00:52:59.730
Faheem Javed: And.

420
00:52:59.730 --> 00:53:12.889
Axcel ILT: Yeah, Fahim, thank you. Thank you so much. And I think Fahim has to teach a class in about 7 min. So thanks for staying on and taking the questions. And I just in case you didn't see in the chat.

421
00:53:13.050 --> 00:53:27.650
Axcel ILT: This session has been recorded, and we'll we'll send you the URL after afterwards. Oh, yeah, for him. It's to go and teach a class he's in demand. Thank you all very much for for joining us. We really appreciate it.

422
00:53:27.990 --> 00:53:43.119
Axcel ILT: And we will be in touch after all. Thanks, everyone. Thank you. I will. I will. I will send it on to behim we really appreciate that you spent this hour with us, and we hope you'll join us in the next one. We'll probably be having another Microsoft, maybe a Microsoft security

423
00:53:43.250 --> 00:53:50.659
Axcel ILT: webinar in in July, and possibly a copilot Microsoft co-pilot in August. So we'll let you know about that

424
00:53:50.880 --> 00:53:55.680
Axcel ILT: alright. Everyone have a wonderful day, and thank you so much again. Bye, all.

Exploring Microsoft Fabric: A Unified Data Engineering and Analytics Platform [Webinar]

Alert!

Modal Title

Error!

Default Title

Prompt

Confirm

Login