View Recording: The Power of Fabric + Purview: Modern Data Governance

0:0:5.411 –> 0:0:21.491 Joe Steiner Hello everyone, I’m Joe Steiner, Solutions Architect here at Concurrency. I want to welcome you today to talk with us about the power of Fabric and Purview, kind of the leveraging the Microsoft tools for modern data governance. 0:0:21.531 –> 0:0:36.491 Joe Steiner So through the course of the day today, we’re going to tour fabric and then we’re going to tour the relevant parts of purview related to fabric and then we’ll kind of talk about how those. 0:0:36.571 –> 0:0:55.411 Joe Steiner Fit together in some of the things that we’re talking to clients about and clients have been working through in terms of how do we govern our data better. This highly relevant right now is we’re exploring, you’ve long had this as you’ve been doing analytics projects. 0:0:56.251 –> 0:1:12.611 Joe Steiner But particularly with the emergence of a I, this has been a a hotter topic because you want to make sure that you’re only surfacing things that people should be able to see when they’re asking a prompting A I and asking questions of it. 0:1:25.288 –> 0:1:42.288 Amy Cousland No, I’m gonna just take a break every second. Somebody’s saying the meeting’s not audible. Can other people hear OK? I’m having no issue hearing on my end. If anybody else could chat in the chat box. I’m not having any issue. Yeah, everybody else is OK. So I’m sorry. OK. Thank you so much. I just wanted to check. I appreciate it. 0:1:30.648 –> 0:1:30.888 Joe Steiner Hmm. 0:1:31.368 –> 0:1:31.648 Joe Steiner Hmm. 0:1:37.408 –> 0:1:38.648 Joe Steiner Yeah, please. 0:1:43.808 –> 0:1:44.488 Amy Cousland OK, go ahead. 0:1:44.688 –> 0:1:48.328 Joe Steiner Yeah. Can every are we? Are we good now? Can others? 0:1:48.848 –> 0:1:54.968 Amy Cousland It looks like nobody else is having issues, so maybe just a matter of their issues up there in settings. So thanks. 0:1:52.208 –> 0:2:2.128 Joe Steiner Got it. Fair enough. Fair enough. No. So yeah, so diving in, then we’ll continue on with. 0:2:3.328 –> 0:2:21.328 Joe Steiner Fabric. So Fabric is a lot of different pieces that have existed. Some of these have existed for some time, some of these are newer. But what Microsoft has done with Fabric is brought all these together in a more user friendly way and in a lot of ways. 0:2:21.888 –> 0:2:41.8 Joe Steiner One of the things I do wanna state up front though is as they’re bringing these things together, sometimes we’ve had some clients say, well, I used to be able to do this with what was in Azure, but now this is a little limited. That’s continuing to grow. It’s not really limited in what can. 0:2:41.568 –> 0:2:56.528 Joe Steiner Be done. It’s just sometimes we still have to use some of the external tools on that too. So for anybody that’s worked with Fabric, there are ways around anything. So anything within the Microsoft Azure framework can be leveraged here. 0:2:57.328 –> 0:3:12.608 Joe Steiner But the thing that’s really fascinating about Fabric is they’re making this easier and easier for people to do that maybe didn’t have the deep technical backgrounds, but do understand what they want out of the data a little more. 0:3:12.688 –> 0:3:29.8 Joe Steiner And so it’s really been a fascinating development there and democratizing what was that, you know, the data science, data engineering plane, not entirely getting away from needing to have those kind of skill sets, but. 0:3:29.88 –> 0:3:45.528 Joe Steiner Making it simpler for people to get things done. So we’ll go through most of these through the course today. Data factory, data engineering is kind of a combination of different things. We’ll talk about it in different ways. Data warehouses, we’ll talk about one lake. 0:3:45.968 –> 0:4:5.768 Joe Steiner Some of the data science tooling, real bit on real time intelligence and Power BI. But really what we want to kind of give you a tour of is there’s a lot of things you can do within Fabric to manage and govern your environment and make it easier to ultimately. 0:4:5.808 –> 0:4:24.808 Joe Steiner Deploy data products and you know analytics workloads and tools that are delivering insights to the organization, but also doing that in a well managed, well governed, secure fashion. And so we’ll be talking about that throughout the the course of this today. 0:4:25.168 –> 0:4:44.88 Joe Steiner So we start with with Fabric. Our topics will be again one lake, one lake’s really foundational to to what is inside of Fabric as kind of linking in how do we storing that data or linking it in as we’ll talk about a little bit. 0:4:44.208 –> 0:5:2.8 Joe Steiner Also I will talk about Data factory and pipelines and what that allows for. Talk about real time intelligence and some of the real time data capabilities within fabric alongside. 0:5:2.208 –> 0:5:18.688 Joe Steiner The more static data platforms that that you may be leveraging, talk about some of the data science tooling, Power BI and then throughout this, there’s a lot of copilot enablement within the fabric workloads as well, so. 0:5:19.288 –> 0:5:33.688 Joe Steiner Things where you can leverage copilot and prompt copilot to do something within, say, Data Factory or Power BI or a couple other areas, and they’re enabling more and more of these all the time. 0:5:34.208 –> 0:5:51.608 Joe Steiner Which makes it then even easier to be able to again work within fabric for your your data workloads. So we’ll we’ll showcase the current state of those and understand that is that is a continuing that that play that space continuing to evolve. 0:5:53.248 –> 0:6:12.328 Joe Steiner So let’s start with One Lake. One Lake is really they will describe this as the OneDrive for data. It’s pretty good at definition where within One Lake I can have multiple different types of data stores all. 0:6:13.8 –> 0:6:29.88 Joe Steiner Principle by all the other things I want to do with the data, the tooling I want to use against the data, whether that’s ingestion, whether that’s data engineering, wrangling, performing data science efforts where I’m. 0:6:29.488 –> 0:6:45.448 Joe Steiner You know, doing some deeper analysis on that, maybe modifying it in a different way, leveraging Power BI against all of that. So it brings all of those what would historically have been disparate data store. 0:6:45.728 –> 0:7:4.328 Joe Steiner Or is I make it easier to to work against all of those. The other nice thing. So with with that you have your one lake catalog which provides that central view of OK, here’s all of our data in the organization that we have tied in with this that’s available. 0:7:4.768 –> 0:7:24.408 Joe Steiner To be consumed, to be analyzed, to be utilized for different purposes. You also have File Explorer here, which is interesting. That can actually integrate in with Windows File Explorer. So from my PC I can look and see if I’m allowed privileges to do this. So for those that are managing. 0:7:24.688 –> 0:7:40.8 Joe Steiner In this environment I can actually see the data items that exist inside of the workspaces and be able to see those in a folder and then a file like structure which shows me that database or that data item inside of there. 0:7:40.368 –> 0:7:55.608 Joe Steiner Like we said, inside of one lake you’ll start with workspaces and workspaces are important. Here’s where we talk a little bit about the governance and security here. I can have separate workspace for, you know, one team or department or use case. 0:7:56.208 –> 0:8:11.728 Joe Steiner Away from another and can secure those so that within that data, only those people within that workspace can do certain things with that data. I can then have a separate workspace where. 0:8:12.848 –> 0:8:28.368 Joe Steiner Only certain people can work within the data within that area, so allows for the beginnings of controlling access, controlling who might be able, particularly for who can modify versus who can view. So you have role based controls that can be applied at that level. 0:8:29.328 –> 0:8:46.768 Joe Steiner And then within there you have your data items, which might be OK. I’ve got a data warehouse in here alongside a a lake house alongside, you know, a standard just SQL database that I’ve that I have maybe being mirrored in there or copied. We’ll talk about that a little bit. 0:8:47.48 –> 0:9:5.688 Joe Steiner So it allows me to group those into manageable units, not only for organization, but also for security and for for governance purposes. One of the nice things that you you have in here aside from the workspaces and then those data items underneath, which we’ll talk a little more about in a moment. 0:9:5.888 –> 0:9:23.128 Joe Steiner We also have the ability to to link in shortcuts. So with shortcuts I can start bringing in maybe on premise sources. I can bring in things from other clouds. So I may have data both in Azure and in Amazon. I can have shortcuts into there so that there is a connection at least. 0:9:23.648 –> 0:9:41.88 Joe Steiner To that other data source, whether you replicate that into one lake or not will depend on performance needs, what you’re doing with the data, those kind of things. Those are areas that we certainly can help you out with making decisions against, but it is a nice another option. 0:9:41.768 –> 0:9:56.768 Joe Steiner Aside from moving data that I can actually link to that data and be able to work against that data, even if it’s outside of being stored inside of the Azure instance underneath one lake, I also can link then. 0:9:57.408 –> 0:9:59.528 Joe Steiner Data items that might be in work. 0:10:1.8 –> 0:10:14.488 Joe Steiner Space B they can do that even if they’re not allowed access into workspace B to modify and and do other things there. So and that’s something you would control and. 0:10:15.808 –> 0:10:33.368 Joe Steiner You can make sure again that your that meets your governance and security concerns, but it allows for for work to happen there without again having to replicate that over into workspace. A allows for a more a broader view of the data that that you have within the organization. 0:10:34.8 –> 0:10:50.248 Joe Steiner Now without actually moving everything right away, let me kind of move the things into here that are gonna be you’re gonna need more performant analysis against querying against reporting against things like that that would be available otherwise so. 0:10:50.448 –> 0:11:5.648 Joe Steiner A lot of a lot of capabilities there that that can be leveraged. As far as storing data in one lake, there are 5 principal. 0:11:6.168 –> 0:11:22.728 Joe Steiner Kind of broader database structures that are possible there. The first one is our event houses. So event houses handled real time. These this will be you know volumes things that you’re getting on a regular ongoing basis. 0:11:22.768 –> 0:11:38.208 Joe Steiner Typically, you know the logs of different kinds be all kinds of of readouts from different systems. If we’re doing things with digital twinning, we might have data in here like that from IoT sources. 0:11:39.128 –> 0:11:55.48 Joe Steiner So all that would be house inside of my workspace inside of one. I also can leverage this is in preview right now, but they’re coming forward. This is Cosmos DB. So here I can do. 0:11:55.808 –> 0:12:1.88 Joe Steiner No SQL data AI. 0:12:2.688 –> 0:12:17.448 Joe Steiner Not as as structured of a data sources as typically not so much a tabular data there that but I I might have want to have a little more flexibility you know start getting into graph data. 0:12:18.88 –> 0:12:33.728 Joe Steiner Bases, things like that that can be stored inside of the the data lake as well. SQL databases. If I just have a single SQL database, frequently I do that for kind of operational data, transactional databases that are used. 0:12:34.208 –> 0:12:48.808 Joe Steiner On a regular ongoing basis. I’m just taking data in there and and so that could be a data item in there also in preview, but it is available and then you have data warehouses or. 0:12:49.648 –> 0:13:6.448 Joe Steiner Or a lake house. One of the biggest differences there is if you need unstructured data, lake house had had been the way to go. Data warehouses don’t aren’t really. Warehouses are more of your traditional data. 0:13:6.528 –> 0:13:22.488 Joe Steiner Warehouses where it’s SQL based structured data on that in inside of there. If I need within that database to have unstructured data and and want to be maybe. 0:13:22.888 –> 0:13:39.648 Joe Steiner And again same workspace. I can have a warehouse within there for that team that’s allowed access to that workspace to operate and that again can be linked off into other teams workspaces too, but whoever owns that data should have that in the workspace. 0:13:41.248 –> 0:13:57.448 Joe Steiner So from there, let’s talk a little bit about, OK, I’ve here, I’ve kind of figured out how I want to have things structured. How am I getting that data into Fabric? And this is where the the data factory which has existed in Azure for some time. 0:13:58.8 –> 0:14:3.368 Joe Steiner But has been built into fabric. 0:14:4.368 –> 0:14:7.648 Joe Steiner And pipelines can be levered. So here it doesn’t matter if I wanna do. 0:14:10.608 –> 0:14:30.128 Joe Steiner There are over 170 connectors that are pre-built there for different types of data sources that I set up my connector. I can either manually trigger that or I can have it set to automatically trigger based on a. 0:14:30.208 –> 0:14:43.488 Joe Steiner Schedule or some event happening. There can be a couple of different ways to to trigger that for it to make that either copy and I can do a copy job I. 0:14:44.128 –> 0:14:50.688 Joe Steiner Can do kind of like the copy or I can do mirroring inside there so that it’s I’m constantly updating. 0:14:51.848 –> 0:14:59.368 Joe Steiner The two sources have again automated pipelines here where I can schedule things. I can have it triggered off of certain events. 0:15:1.88 –> 0:15:20.168 Joe Steiner It’s automatic into my my data item inside of one lake and making it available for for to be able to be worked on. The other thing you’ll have in there is data flow Gen. 2 which allows for again that a different version of that automation. 0:15:20.688 –> 0:15:36.648 Joe Steiner In an ingestion there, it’s built into to Fabric, and here’s one of the the first place where we can start applying copilot where there are certain prompts I can make to say, hey, set up a. 0:15:37.328 –> 0:15:56.488 Joe Steiner Connection to this database on this schedule and it can do that. It’ll take care of implementing that for you within Fabric and within the using the data factory tooling therein. So very very powerful. It makes it again easy. 0:15:57.168 –> 0:16:0.328 Joe Steiner Easier and easier to to leverage. 0:16:1.648 –> 0:16:2.8 Joe Steiner Of work. 0:16:4.88 –> 0:16:11.88 Joe Steiner Just instant data, but also people that may not be as as as skilled in those areas. 0:16:13.168 –> 0:16:31.848 Joe Steiner So we talked that there is some real-time intelligence capabilities here. Let’s let’s talk about that for a moment. The kind of the foundation for this. Again, the storage of this is in the event house, but within fabric there is a a thing called the real-time hub which allows me to. 0:16:32.208 –> 0:16:47.568 Joe Steiner Have control over the the the full range because real-time data is a little different than kind of your traditional data sources. It allows me to control all those specific tools to be able to ingest. 0:16:47.648 –> 0:16:59.48 Joe Steiner Real time data, which would be inside of Event Stream, store it in Event House, can use the activator to actually act on data. So that might be a notification, it might be triggering. 0:17:0.448 –> 0:17:17.888 Joe Steiner Some kind of automation that occurs if we see certain events happening inside of the real-time data that can be built inside of Fabric and triggered. I have my ability to have a real-time dashboard because that data is constantly changing. That dashboard’s a little different than. 0:17:18.8 –> 0:17:34.168 Joe Steiner Perhaps dashboards I built off of traditional data sources. I also have a digital twin builder inside of here. So again for those IoT workloads where I’m building a digital model of perhaps my machines on my factory floor or. 0:17:35.408 –> 0:17:50.968 Joe Steiner Of different pieces of equipment inside of my facility, I can build that digital twin inside of the real time hub in Microsoft Fabric. It is capable of leveraging the Kusto query language. 0:17:51.328 –> 0:18:8.248 Joe Steiner Very useful for a real time data and again all this is the data stored within one lake within all of alongside all my other data so that as I’m generating say Power BI reports, I can be incorporating real time data into there alongside. 0:18:8.808 –> 0:18:11.768 Joe Steiner My more traditional data sets. 0:18:13.168 –> 0:18:30.48 Joe Steiner So inside of Fabric we then you know aside from the storage and ingestion of that and maybe along with some you know wrangling I want to do some things that are a little more sophisticated than the automated. 0:18:30.288 –> 0:18:48.728 Joe Steiner Kind of core trend transformations that that I’ll be doing with the data. I can build notebooks inside of Fabric that work against the same data sets, leveraging all the other tooling. I can run experiments inside of there. I can establish environments for. 0:18:50.8 –> 0:19:6.448 Joe Steiner Different use cases that I that I may have and be able to you know again if I’ve got this problem I’m trying to solve and I want to run my experience, I can do my data discovery processing inside of Fabric. 0:19:7.88 –> 0:19:8.928 Joe Steiner I can experiment in my. 0:19:10.328 –> 0:19:11.248 Joe Steiner On that the. 0:19:16.168 –> 0:19:32.368 Joe Steiner An operational product or model or automated workflow that maybe that gets incorporated into the ingestion process or into the reporting end or wherever I might be leveraging that but. 0:19:32.448 –> 0:19:37.128 Joe Steiner It allows me to experiment inside of Fabric in a. 0:19:38.848 –> 0:19:55.408 Joe Steiner Safeway, but it gives me the tools to be able to do that. Couple of the interesting things inside of here, aside from the notebooks and experiments and being able to obviously create models, models not only data models but also machine learning models. 0:19:55.528 –> 0:20:10.808 Joe Steiner Side of here I can use ML flowing inside of here to create models against like that. I have that out there for other purposes. There’s the Data Wrangler tool which can be very useful for handling some of these. I don’t have to. 0:20:11.8 –> 0:20:29.448 Joe Steiner Necessarily build my own custom code to the same degree. There’s a lot of things Data Wrangler can do. The other thing is that you have data agents which really start getting into a starting to leverage a little bit of of AI to be able to process and and wrangle and handle the data. 0:20:30.328 –> 0:20:45.88 Joe Steiner And I can kind of tell it what I want it to do and it allow it to to manage that that for me again making this easier for those that maybe don’t have the the same skill set which it’s kind of the general theme with all. 0:20:45.88 –> 0:20:58.648 Joe Steiner Fabric is how do we again democratize the data state a little bit to allow people to ask questions of it that they’d be able to derive value from even if they don’t necessarily have the data science skill set. 0:20:59.968 –> 0:21:18.248 Joe Steiner There’s a host of the what had historically been the cognitive functions inside of Azure that are my notebook and have it run sentiment analysis against data that I may have. 0:21:18.688 –> 0:21:33.608 Joe Steiner If I’ve got a lot of text data in there, I can have a function that that just says OK run sentiment analysis against this and it’ll kind of give you you know the the data set on that then with a a readout on on sentiment to a host of different. 0:21:33.648 –> 0:21:50.48 Joe Steiner A I functions that can be called from within the notebooks and then be leveraged from there. They’re also again, here’s another place where they’ve enabled copilot to be able to again just establish prompts and say, hey, I I want to be able to do this. 0:21:50.528 –> 0:22:9.208 Joe Steiner Can you can you help me with this? And it can decide the best way to handle what you’re asking and it will kind of prompt you along for for doing that. But that way you don’t have to write the code, you can just tell it what you want it to do, particularly if you copilot. In these cases frequently you do have to kind of know what you what you want it to do. 0:22:9.728 –> 0:22:24.768 Joe Steiner But you can just tell it you don’t have to write the code as is really the the biggest benefit for copilot in data science there and really you know data factory to some level too. So but it’s just again the whole thing’s about making this whole thing easier and. 0:22:25.168 –> 0:22:40.648 Joe Steiner Allow you to spend time on the more value-added activities that are that are part of this this whole process. Power BI deeply incorporated in here to be a tool to be able to visualize and analyze your data, share it with others. 0:22:41.248 –> 0:22:57.408 Joe Steiner I can bet in other tools like teams as you see in this example here and I’m able to you know share that without the throughout the organization. But I can build these inside of data and leverage the data in one lake and have a a. 0:22:57.408 –> 0:23:17.168 Joe Steiner A robust, secure connection there. Again, I can tie in copilot here to be able to ask questions of the data and this really is where it really opens that up to the to the public within your enterprise a little more because I can have, you know, my executives asking questions directly to the data, especially if I. 0:23:17.248 –> 0:23:36.8 Joe Steiner I’ve done a good job organizing the data, which is really where the government’s going to come in here, as we’ll see in a second. But say, hey, what were the sales for this area over the last three years? And it can, it can answer that for you, leveraging the data and understand as long as I’ve got a good semantic model built on top of that. 0:23:36.248 –> 0:23:51.528 Joe Steiner So we’ve talked a little bit about where Copilot fits in in Fabric in different places. Again, Data Factory, particularly with data flows, Gen. 2 specifically and data pipeline. 0:23:51.968 –> 0:24:11.368 Joe Steiner Data science, I can be leveraging Copilot within the notebooks. Data warehousing. There are some capabilities as I’m managing my data warehouse to be able to do that. The data warehouse specifically Power BI again, creating the reporting, even creating the semantic models. I can leverage Copilot. 0:24:11.568 –> 0:24:26.448 Joe Steiner To help with that SQL databases and then in the real time space I can use Copilot to write KQL queries and to operate with the real time dashboard. Very similar to how I’d be operating with reporting in Power BI. 0:24:27.368 –> 0:24:45.768 Joe Steiner Just to be on the real-time dashboards customized for for real-time workloads. So I expect you’ll see growth on this screen that these are the nine areas right now that that they’ve already built these for, but that’s continuing to grow. There’s definitely investment here. 0:24:47.8 –> 0:25:2.808 Joe Steiner So you will, you will see this expand over time again, being able to leverage that in the power of, you know, even your voice with copilot and telling it what you want it to do, dictating to the machine, hey, you know, computer do the copilot, take care of this for me. 0:25:3.648 –> 0:25:20.928 Joe Steiner Please and and let it run. So very, very powerful. But as I’m doing that, I want to make sure I have a strong foundation. Again, copilot, whether it’s in fabric or copilot in general where I’ve maybe exposed some data. 0:25:21.488 –> 0:25:36.608 Joe Steiner To be available to ask questions of it within copilot more broadly, I want to make sure that I’ve got a strong foundation here. One, I need to have my data organized. I need to have it secured so that I’m not surfacing data to people that shouldn’t see it. 0:25:36.928 –> 0:25:52.488 Joe Steiner And I want to make sure that as I am surfacing data, even the people that should be able to see it, that that is still tagged and marked as proprietary data so that they don’t accidentally share that off as well, so. 0:25:52.848 –> 0:26:7.808 Joe Steiner A very, very important piece of building out AI inside of the enterprise. So to solve that Microsoft Purview is really the best, the best choice and some of the three most relevant areas for a little bit is. 0:26:8.128 –> 0:26:24.968 Joe Steiner It using data map and unified catalog. These two work together and we’ll talk about those in just a second. It allows you to map your data environment. You can let it crawl over the your different data sources in the enterprise. It doesn’t just have to be within. 0:26:25.328 –> 0:26:42.128 Joe Steiner Data lake can be on premise, it can be Azure, could even be into Amazon and be able to get a sense of what it what do I have out there in terms of data? Maybe does this data exist somewhere? And what this does is pulls the metadata. 0:26:42.528 –> 0:26:57.728 Joe Steiner From those data sets, it’s not pulling the underlying data itself, but it’s pulling the information about what fields are in there, what’s the name of the database, name of the tables, that kind of thing. Being able to see that in a unified catalog then. 0:26:58.448 –> 0:27:15.128 Joe Steiner Along with being able to see the permissions on there and any any controls there and sensitivity labels and what have you. Sensitivity labels then falls into information protection, so the information protection form part of Purview. 0:27:15.568 –> 0:27:32.768 Joe Steiner Purview has a Microsoft 365 side where I maybe are putting sensitivity labels on my SharePoint, OneDrive, Teams data looking for certain types of information sensitive that might be sensitive to the person. 0:27:33.288 –> 0:27:50.8 Joe Steiner Sensitive financially, sensitive for health reasons. It can search for HIPAA and all those, but I also may have my own sensitivity labels within the corporation. So anything related to this, you know, code word, this project name, this. 0:27:50.288 –> 0:28:9.768 Joe Steiner Perhaps enterprise that we’re acquiring that not everyone should know about financial data that a proprietary organization, trade secrets, all those I can set up data sensitivity labeling inside of there in SharePoint. 0:28:10.288 –> 0:28:29.448 Joe Steiner Point and those tooling, but I also can apply that to my structured data and so I can use purview within structured data, particularly within fabric and be able to tag certain data sources as more sensitive so that maybe I don’t want to expose. 0:28:29.728 –> 0:28:45.168 Joe Steiner This financial data to everyone, only certain people get to see it. So kind of the way that will work is I label it and then choose what protection policies I want to have on that, any built-in encryption there. I also then can leverage that against. 0:28:45.768 –> 0:29:2.408 Joe Steiner My DLP engines, whether I’m using Defender, whether that might be in some of the Microsoft 365 workloads, but they’ll also be recognized by within Copilot as well and other cloud apps if I have those those tools built up that way. 0:29:2.688 –> 0:29:17.608 Joe Steiner So that to prevent that data from being shared where it shouldn’t be. Finally, audit’s a very important piece of all this to be able to track what is happening inside the environment, who’s accessing what, ensuring that you have things set up the way that you want to. 0:29:19.168 –> 0:29:33.568 Joe Steiner So for data mapping and and unified catalog, once I’ve had the data mapping tool run against my environment and go into the catalog and search against any of the available metadata there, so. 0:29:34.288 –> 0:29:51.448 Joe Steiner Let’s say I wanted to search for any of the sales data I might have there. It’ll show me, OK, you’ve got this Power BI data set, you have this report, you have certain things related to sales in BLOB storage. You’ll note here that some of these have sales. 0:29:51.888 –> 0:30:9.608 Joe Steiner In the name of the file that’s that’s stored there. Some of this might be because there is a label and or it’s in a certain folder. So it’ll take all that that data and be able to show me OK, where might I have my sales data across all of this? 0:30:10.168 –> 0:30:28.48 Joe Steiner As I leverage tagging and we’ll talk about domains in a little bit and that this really gets to be even even better and those search results are prioritized when I come into here to search for the type of of data that I that I have out there. So it’s a great way to. 0:30:28.168 –> 0:30:47.208 Joe Steiner Kind of map out my data environment and then I can start working at how do I want to manage this and make it easier to use, but at the same time as I’m doing that to make sure that I’m not oversharing and or making it too easy for the wrong people to be able to access this information. 0:30:47.648 –> 0:31:4.568 Joe Steiner So the next part of that then is then information protection. And again, the first part of that is knowing your data. So map it, study it, and then along the journey of protecting it, this is where you start putting labeling on different data structures. 0:31:5.848 –> 0:31:24.928 Joe Steiner Being able to to map that out, I may not put the protection on 1st until I’ve gotten my labeling, my automated labeling set up the way that I want, and then once I’m comfortable with that, I can start putting in measures to protect it and prevent data loss by, you know, maybe encrypting that data. 0:31:25.8 –> 0:31:43.808 Joe Steiner So that wherever it goes, it remains encrypted and or preventing other people from accessing it. Setting permissions on. If you can’t view this this type of sensitivity label, even though I could view the data source, it will stop you from being able to see that that line of data or. 0:31:43.888 –> 0:32:1.568 Joe Steiner Things like that can be used for those purposes. And then you know, I want to continually govern this and make sure that I’m, you know, managing this over time while still making this as easy to use as possible. You want to be able to. 0:32:1.688 –> 0:32:20.128 Joe Steiner This is how you can securely say yes to be able to allow people to ask questions of the data really is having a strong information protection there. So you know, obviously enabling persistent data, data protection, so applying sensitivity labels. 0:32:20.408 –> 0:32:36.288 Joe Steiner Ensuring that the classification is appropriate for what’s what’s in there. But this again, those data labels travel with the data. So even if I’m just labeling it, I can then see where that the data maybe is is going, where it’s been shared out. 0:32:36.568 –> 0:32:55.168 Joe Steiner And and maybe what’s happening with that then can I automate governance here where as I have that label applied, I can apply protection policies that will automatically enforce certain things as in if I’m not within the appropriate security group, I can’t access this data. 0:32:55.688 –> 0:33:15.168 Joe Steiner Or if I was in that security group and I no longer am, even though I have some of that data, I I can’t access it anymore because I’m no longer permissioned to be able to do that, even though nothing has changed in terms of me receiving the data or data moving, it’s just I have it on a on a store. 0:33:15.368 –> 0:33:35.48 Joe Steiner Or I no longer can see it because my permissions have changed. Could be because I’ve left the company, could be that I’ve changed roles. There could be a host of reasons for that. Allows you to simplify compliance too. So here I can actually have with the data classification on there. I can have an auditable trail of where data’s going, which can be very. 0:33:35.48 –> 0:33:49.808 Joe Steiner Useful if if there are any questions in terms of compliance. There allows you to collaborate with with this data without having to worry so much if you’ve done a good job of putting in the built in protections. 0:33:50.328 –> 0:34:7.648 Joe Steiner From one end user to the other within your organization and say, hey, I want to show you this database. It will permission them so they can only see what they’re permitted to see. You’re not. You don’t have to worry about people accidentally sharing things that will give them visibility that they shouldn’t have and so. 0:34:7.848 –> 0:34:25.968 Joe Steiner If you have a a strong data protection plan and policy in place and have the protections in there, that’ll that’ll help prevent that from happening and then just allows me to centralize all this. I can control all this from within purview so. 0:34:26.328 –> 0:34:44.488 Joe Steiner There is a purview hub we’ll show you in a moment that’s built into fabric and so from within there I can see, all right, where is my data, where is my sensitive data, what’s happening with that? And maybe, you know, recraft your policies or manage your policies from there. 0:34:44.928 –> 0:35:0.168 Joe Steiner As far as DLP, which is kind of that next layer of protection where I can’t, you know, get to it, I can’t share it outside of just having the the labeling on there, I can establish policies there. 0:35:0.168 –> 0:35:15.768 Joe Steiner And that will stop me from accessing that data or moving that data for certain types of information. So even if I am able to view it myself, this will stop me from moving it around and accidentally putting it somewhere where it shouldn’t be. 0:35:16.648 –> 0:35:32.488 Joe Steiner That allows me to scope those to certain fabric workspaces. So for example, if I have a workspace for the finance group, I might just put a policy on everything within that finance workspace and that applies to all the then data items underneath that. 0:35:33.128 –> 0:35:49.528 Joe Steiner So that that helps me just protect that more broadly and say, OK, anything finance is working on, we don’t necessarily want everyone else to be able to just to see until the right time. There can be regulatory reasons for that certainly. So I can consider. 0:35:49.528 –> 0:36:6.88 Joe Steiner Create actionable alerts, notifications. So if there are policy violations, things have happened. I can notify somebody and you can handle this through either education and or taking action. You can see what actually happened in real time. It also allows users to be notified that hey. 0:36:6.808 –> 0:36:26.368 Joe Steiner What you’re trying to do isn’t permitted and be able to stop them from taking an action. And again, all this again is audited. So this it’s again important to be reviewing your your audit logs with all of this in order to make sure that what you’ve set up is working overtime and that you don’t have other exceptions you need. 0:36:26.488 –> 0:36:45.888 Joe Steiner We need to take care of and or maybe need to change the policies because we’re blocking a lot of stuff that maybe we shouldn’t. So it is, it is something that needs to be managed over time. So we talked a little bit about power, fabric and purview together and how these can be be leveraged, but let’s talk about it in kind of more general terms. 0:36:46.248 –> 0:37:1.568 Joe Steiner In terms of, you know, if if I’m looking at my data set, what am I really trying to accomplish out of these things? Well, you know, one of the things I want to do with between Fabric and Purview is organize and curate my data. This allows it to make it more findable. 0:37:2.328 –> 0:37:18.488 Joe Steiner Easier to discover, allows people not have to search so much for where data might be as they’re trying to do different kinds of analysis or build certain data products. It allows it to to bring it all in one place. So how do I leverage Purview and Fabric to help with that? 0:37:18.568 –> 0:37:35.48 Joe Steiner Also then, as we’ve been just talking about recently, how do we secure and protect it, right? I made it discoverable. How do I make sure that it’s not too discoverable for the wrong people, that only those people that should be able to access that are are able to. So how do I leverage Fabric and Purview together? 0:37:35.528 –> 0:37:55.88 Joe Steiner To be able to secure and protect my data. And then you know, in addition to that, so I’ve got it, I’ve got it discoverable, I’ve limited what people can see inside of there to the right levels. But then over time, how do I ensure that people have confidence in that data? And this really gets into. 0:37:55.288 –> 0:38:14.768 Joe Steiner How am I providing some kind of validation that yes, this is good data, the company believes in it, this is the best data source to use, and there’s some tools inside of Fabric and Purview to help promote that to track lineage of data and things like that in order to. 0:38:15.168 –> 0:38:33.448 Joe Steiner Ensure that people have confidence they’d now be able to find it. They can only see what they should be able to see and there are limitations on where that can go from there. And I’m promoting confidence in that over time so that people are confident in that. OK, I found this data and I’m comfortable using it for this purpose and so you’re not. 0:38:33.528 –> 0:38:48.528 Joe Steiner Spending the cycles questioning that to the same degree. And then finally, you know, the whole purpose of all this is to be able to use the data. Just having the data is one thing, but I want to be able to use it, and so I want to be able to drive insights for that. I want to be able to. 0:38:49.128 –> 0:39:6.8 Joe Steiner Build applications and data products off of that. I want to be able to leverage that for AI use cases. So all of this is very important to be able to get to that point. So in terms of organizing, curating your data, some things we’ve talked about already, purview data map. 0:39:6.568 –> 0:39:21.408 Joe Steiner A unified catalog. You know, you start there by mapping and catalog your data across multiple clouds and platforms. Gives me kind of a sense. OK, what is my data state actually, right? It’s not just what’s in this bucket and what’s this in this bucket. It kind of can take a view across. 0:39:21.608 –> 0:39:39.368 Joe Steiner Many different ways if you if you set that up the right way and be able to to start bringing that all in and start be able to think about OK, how should I be managing this a little differently? Way to start managing that is OK, how am I linking that into one lake so I can put my trusted data sources and any data source I want to have in there. 0:39:39.448 –> 0:39:55.408 Joe Steiner Be able to LinkedIn there, whether I’m migrating that data source into there and then eliminating another data structure that I have to manage, I might be copying that in so that data has to continue to sit where it is, but. 0:39:55.768 –> 0:40:12.648 Joe Steiner I’m now copying that into one lake, mirroring that. So if it’s it happens on a more regular basis or still like, hey, that’s going to sit out there. I don’t really want to port all that in, but it would be nice to be able to interact with it at some levels, even if it’s not going to be as performance. 0:40:12.888 –> 0:40:29.688 Joe Steiner And that’s where I can leverage linking the shortcuts. So once I’ve kind of mapped out where my data all is everywhere, I can start leveraging one lake to be able to start consolidating that all together, making it easier for the end users and for those involved with managing the estate. 0:40:29.688 –> 0:40:44.448 Joe Steiner To be able to to see what and control what’s what’s in there. From there I can set up the concept of domains. So here across different workspaces and data structures I can set up. 0:40:45.528 –> 0:41:5.448 Joe Steiner What I might have a business domain, so I might have a finance domain. Finance may have multiple workspaces out there and they might, you know, have some other outside data sources there. I can establish a domain across those to say, OK, this is all the finance domain data and from there then that can be useful for the next. 0:41:5.528 –> 0:41:25.88 Joe Steiner Piece when I’m starting to secure or protect this data about OK, I can start to manage that at a domain level across all those pieces of my data state. The other thing I can leverage here is tagging. Tagging provides that additional metadata where if the name of a table or a file. 0:41:25.528 –> 0:41:44.648 Joe Steiner File or a database doesn’t necessarily convey where this might be useful from a business sense. I can put tags on data too so that that adds that extra metadata that then is consumed by the data map and into the unified catalog so that as I’m searching for data I can see wait a minute, this is sales related. 0:41:44.688 –> 0:42:3.928 Joe Steiner Sales related to even it says nothing about sales. This allows me to put that sales tag on there. I mean I can even get far more granular than that, that this is sales for this division versus that and be able to differentiate especially when maybe their underlying sales data looks the same, but they’re for different divisions. 0:42:4.208 –> 0:42:23.448 Joe Steiner And use tagging to be able to differentiate that way. Very, very useful for again organizing and starting to curate the data to make it easy to find the data that I want. Now I’ve made it easy to find. I have to make sure that I’m again haven’t made it too easy and so I need to make sure that I’m securing this the right way. 0:42:24.248 –> 0:42:42.848 Joe Steiner 1st place to start is if I’m doing things within fabric, either be between the tenant and the workspace, I want to make sure I’ve got the right rule based access controls on there to be able to control access into that fabric environment in the right way. Make sure people aren’t able to alter that data, make changes to it, extract it. 0:42:44.168 –> 0:43:3.248 Joe Steiner When they when they shouldn’t be able to at a base level. From there I can start leveraging sensitivity labels. Well, maybe I can access this data, but I want to make sure I’m careful about where that data is going over time. And so I can leverage sensitivity labels to say, OK, here’s some data we need to be a little more careful about which then. 0:43:3.568 –> 0:43:18.728 Joe Steiner We can have the protection policies and DLP act against where if that’s starting to be exfiltrated anywhere, that can put a stop at it and be able to track what’s happening there. And again, all that will be audited so that I can use the purview audit to be able to track for any. 0:43:18.928 –> 0:43:33.848 Joe Steiner Ongoing activity, maybe areas that I hadn’t put the protections on in the way that I maybe had wanted to or should have what have you, but I’m able to to see what’s happening there and I can then alter my management course along that. 0:43:34.408 –> 0:43:50.728 Joe Steiner From there I’ve organized it, curated it, secured it. How do I now promote confidence in the data? And this is where the concepts of endorsements come in. So here I can endorse the quality of data contained in certain data items. So within. 0:43:51.648 –> 0:44:10.848 Joe Steiner I might say, hey, I’ve got this data warehouse versus this. I’m doing a bit of a medallion architecture as some of you may be familiar with. I can actually put an endorsement on there that then is carried through onto Power BI into anywhere where more end users will be working on it. You can have. 0:44:11.168 –> 0:44:28.688 Joe Steiner Them work against only data that’s been endorsed in a certain fashion allows to again, you’re promoting trust here and promoting confidence in the in the data set. You may have more raw data underneath that has been modified and you keep the raw data there for other purposes. 0:44:29.8 –> 0:44:45.648 Joe Steiner You don’t want everyone operating against that. You want everyone offer operating against that more curated, managed, trusted data. And this is a way to to tag that essentially and be able to say, OK, this is the data source you want to work against. Even though you see 3 copies of this, this is the one to use. 0:44:45.728 –> 0:45:4.968 Joe Steiner I also can have lineage tracking inside of here built in which allows me to see where did this data come from and so I can see that this this data was transformed and extracted from here. It’s then put into to this structure and then I. 0:45:5.448 –> 0:45:24.768 Joe Steiner These fields out of there and these fields out of here to create this table or this data structure here. It’s important because I can see where the data came from, so I understand perhaps more context of what that data might be representing. The other thing I can do that tied in with lineage is what is the impact in that? 0:45:24.848 –> 0:45:44.408 Joe Steiner Analysis. So if I am making perhaps looking at making modifications to a data structure that might have be further up the lineage for another data structure, I can see what’s tied to this so that where might that those impacts, what might be those impacts be going downstream. 0:45:44.968 –> 0:46:2.88 Joe Steiner How is that going to impact others when I make this change? And so there’s a impact analysis capability built into here too. All of this is visible within the Purview hub for Fabric. There’s a kind of quick glimpse of the front page. It’s very small, but you’ll see that there are. 0:46:2.368 –> 0:46:21.768 Joe Steiner Sensitivity labels there, there’s the the endorsements, there’s domains. So I can view my data state from different points of view there, be able to see lineage through here. I can see the impact analysis on there and get. 0:46:21.888 –> 0:46:41.648 Joe Steiner Down into the data item level, so the specific data warehouses and that and be able to see different data points on there. I can see how widely used are my sensitivity labels, how have I put on endorsements onto a number of structures or is that pretty limited right now? 0:46:41.688 –> 0:46:58.768 Joe Steiner Now maybe do I need to think about focusing on that? And how much am I using domains across the data state there? So just as some quick examples, there’s a lot that can be done inside of this space, but very useful. And again, this is Purview Hub is built into Fabric. It’s one of the. 0:46:58.848 –> 0:47:1.568 Joe Steiner Tabs you’ll see inside of Fabric. 0:47:3.528 –> 0:47:20.88 Joe Steiner So finally, you know, I want to be able to use this data. How am I doing that? So you know, frequently it’s going to be Power BI, maybe the fabric data science tools, maybe some other data science tools, copilot, but within those Microsoft controlled. 0:47:20.968 –> 0:47:36.768 Joe Steiner Tool sets tied to Fabric. The work I did to organize and curate my data allows me with unified catalog domains and tags has made it easier to for my end users to find that data to focus on the the higher quality data and. 0:47:37.168 –> 0:47:56.768 Joe Steiner And be able to get to work faster. I’m not searching for data, so I haven’t made the search process so difficult. I then have also if I’ve secured and protected the data with sensitivity labels, protection policies and DLP, those are respected in these tools. So if I. 0:47:56.768 –> 0:48:10.488 Joe Steiner I’m in a Power BI dashboard. I’m logged in as myself. It matches my identity against what I can see. If I’m only allowed to see North American sales, but global sales are in there, I will only that’ll filter that view down to. 0:48:11.808 –> 0:48:31.248 Joe Steiner To North American sales from me based on my identity. If I have those labels and protection policies set up the right way, I may be scoping that in other ways, which is fine, but this way ensures me from as I’m setting up new ones, I’ve got that built into the data so that anytime somebody’s viewing the data that’s already pre built into there which makes. 0:48:31.568 –> 0:48:47.8 Joe Steiner Get a little easier again to prevent oversharing and data leakage and to make it easier for people to consume the data in a trusted fashion. Finally, I’m working at promoting confidence in the data, so I found it. I know I can access this. 0:48:47.8 –> 0:49:5.8 Joe Steiner Am I? Do I, as the end user, trust the data? Is this data of a of a high enough quality for me? Is this the latest? Is this data that’s been managed in a way that I would expect on the back end? Or is this a little too raw for my purposes? Am I interpreting this the right way? 0:49:6.128 –> 0:49:25.328 Joe Steiner That’s where endorsements really, really help with saying this is this data, this is what you you should expect that this is. And then having the lineage and ultimately the impact analysis inside of there too to help manage that over time, ensure that you maintain trust in the data so that people are actually able to use it. 0:49:26.328 –> 0:49:40.648 Joe Steiner Is the whole point of having it in the 1st place. So from here it covered a lot in kind of a tour of purview and fabric. Lot of areas you can dive in inside of here. 0:49:40.928 –> 0:49:56.128 Joe Steiner Happy to help you out with that. This is something we’re doing with customers all the time is navigating the space. Want to give you kind of a quick tour of how these these tools work together today. So happy to work with you on a data environment review where we’d be. 0:49:56.568 –> 0:50:14.928 Joe Steiner Happy to identify opportunities to start unifying, managing this in a different way, optimizing your data landscape, securing it perhaps differently and or taking the next step and saying OK, how can I leverage Fabric and Purview together to streamline compliance and make sure that I I can trust. 0:50:15.8 –> 0:50:30.288 Joe Steiner Data and being able to trust the process of building out data insights on off of this. So with that, I want to thank everybody for your time today. I hope this was useful and if there’s any questions, we’ll we’ll take them now. 0:50:40.288 –> 0:50:47.568 Amy Cousland I’m not seeing any questions, so feel free to oh, wait, questions in the Q&A window here. Let’s go ahead and look at those. 0:50:47.968 –> 0:50:51.888 Joe Steiner Yeah, sorry, sorry if I wasn’t. I’ll I’ll pull that up myself too. 0:50:57.248 –> 0:51:16.768 Joe Steiner Yeah. So I’m I’m just going to go from the one that I saw first year. So what’s an example of an impact? How does impact analysis represent the stakes of the impact? How is it? No, I wouldn’t say that it would say the the stakes. I mean I think that’s you know in kind of more the the human terms of that what it can do is. 0:51:16.768 –> 0:51:34.8 Joe Steiner Say that hey, this data field is being consumed by these other data structures inside of there and that if you’re making a change to this, it’s going to be it’s going to be consumed by this other data structure and or data item out there and so. 0:51:34.208 –> 0:51:49.648 Joe Steiner It would just show you where those potential dependencies are between there. As far as the stakes of that, it doesn’t necessarily mean that I make this change. That’s a bad thing. It’s just being aware of where those potential impacts are, so. 0:51:49.688 –> 0:52:6.128 Joe Steiner That’s really what they they’re they mean by the the impact analysis is how widely used is this data set downstream. And so that that gives you a kind of a sense of that as far as is there a systematic way to identify what data deserves to be? 0:52:6.368 –> 0:52:23.8 Joe Steiner Promoted certified master, I don’t know that there’s a there is perhaps a little bit of art to that, but if if we’re pursuing A medallion architecture, there is a little bit there that we could we we could leverage you know again the promoted certified master. 0:52:23.208 –> 0:52:38.368 Joe Steiner Is are the tags that are available for endorsements by Microsoft and they have certain meanings ascribed to those. And so I, you know, we can work with you and say, OK, this is the that definition of what that meaning’s supposed to represent. 0:52:39.128 –> 0:52:56.888 Joe Steiner And that’s really what what what that is. It a little easier in a medallion architecture there to do that, perhaps a little more systematically there we could if I’m kind of building my way to that. 0:52:57.448 –> 0:53:15.248 Joe Steiner That might that might have a little more of a one-off approach to that and or a general policy towards OK where where how comfortable are we with endorsing this as this is this is one or the other if I’m endorsing it at all. 0:53:15.488 –> 0:53:30.888 Joe Steiner So any purview functions that innately key off of that. As far as purview functions that innately key off of that, the endorsement really shows up. It’s not so and so when. 0:53:30.968 –> 0:53:46.648 Joe Steiner It’s surfaced inside a purview for sure. I don’t know that they’re the protections necessarily get driven off of that. They’re more driven off of sensitivity labels and things like that. 0:53:46.648 –> 0:54:3.568 Joe Steiner But it does surface itself. So for example in the unified catalog it would surface those and say OK, if I’ve got master data here, I’m going to show that first that’s it prioritizes in search results when I’m looking at things that way. 0:54:3.768 –> 0:54:22.928 Joe Steiner And that’s and then in Power BI when I’m looking at different data sources there, if I’m looking at crafting A dashboard or report, I will see it there. And I also could if I’m doing perhaps things with copilot, I can say hey only use. 0:54:23.288 –> 0:54:32.648 Joe Steiner Master data, or at least endorsed data at some levels there I could I could make some some changes there. 0:54:34.848 –> 0:54:50.8 Joe Steiner So yeah, it it is, it is kind of tagging those data sources as for a different form of filter, but it does natively show up in the search results. So the the search algorithms designed to be able to promote. 0:54:50.848 –> 0:55:9.488 Joe Steiner Endorse data above others. So there are some built-in things there, but not necessarily in the in the protections. That’s more in the kind of usability side and governance side of of things there. So you have exported a set of data to Excel or CSV in those. 0:55:9.968 –> 0:55:28.88 Joe Steiner Security tags are in place. I’ve accessed revoke later. By what means am I preventing from reading the contents of my exported file? So great question. So if I have an Excel file and the security tags have been carried forward into that as if I’ve set things up the right way. 0:55:30.848 –> 0:55:44.488 Joe Steiner I showed if it came from a managed data source, a protected data source that then I can apply encryption along with the tag. So one I can just tag it so I know what it is and then I can use DLP to stop it from going anywhere. 0:55:44.568 –> 0:56:4.248 Joe Steiner But if I’ve enforced encryption information protection policies on it, I actually can encrypt that file so that if I can’t log into the file, if it can’t identify me or it’s not a trusted login anymore, that that file just becomes an encrypted BLOB, so I won’t be able to get into the file. 0:56:4.248 –> 0:56:18.928 Joe Steiner Anymore. And that’s that’s what those information protection policies do alongside sensitivity labeling. The nice thing with it is sensitivity labels let you identify it first, make sure you get those set up the right way so you’re not. 0:56:19.408 –> 0:56:34.648 Joe Steiner Over enforcing encryption and bringing the business to its knees. So there’s that it’s kind of a nice two-step process there as as you go through that as far as preventing from exporting to Excel compatible formats at all. 0:56:36.88 –> 0:56:50.728 Joe Steiner That will depend on on how you’re setting up policies. That really becomes more a function of some of the the DLP pieces and how you’ve set the controls within both the tenant and workspaces and and things like that. 0:56:51.928 –> 0:57:6.848 Joe Steiner So, but the nice thing with Excel is it is a managed form inside of using purview and information protection that I can enforce security labels and then the information protection policies on it. 0:57:8.408 –> 0:57:22.208 Joe Steiner So if I, yeah, copilot is a decent amount of read-only ability with what tools can it make agentic action and what can it do? So yeah, I I think on that one of the slides we had here, I’ll bring it back up again. 0:57:24.328 –> 0:57:40.448 Joe Steiner This is the current list within Fabric. Obviously Copilot can query different things, but built into Fabric and into the tooling. Here are the specific things I can I can do inside of there or specific. 0:57:40.648 –> 0:57:57.168 Joe Steiner Structures that I can leverage Copilot immediately. Again some of these like in notebooks I can pull up Copilot in the window and it will be operating within the the notebook alongside me. Power BI same way where I can go to use Copilot and. 0:57:57.608 –> 0:58:13.368 Joe Steiner Do natural language querying against it with then that understanding that copilot brings to it. You’ve had natural language querying in Power BI before. copilot allows for a different level of intelligence. 0:58:13.368 –> 0:58:27.288 Joe Steiner To ascertain what you’re asking and what’s available to it and be able to discern, OK, this is likely what you’re looking for and and here you go. Having good semantic models really helps with that. Very, very important. 0:58:29.488 –> 0:58:29.808 Joe Steiner Um. 0:58:32.728 –> 0:58:35.168 Joe Steiner Yeah, it’s uh, let’s see. 0:58:37.608 –> 0:58:57.168 Joe Steiner Is a license required to use the purview functions in in Fabric? So the purview functions in Fabric are run on a consumption basis for the most part. If I have say E5 or that information protection plan to in Microsoft 360. 0:58:57.448 –> 0:59:13.608 Joe Steiner That lets me tag and set things up and automatically apply security labels in my unstructured data. So my office files, PDFs, all those things I might have stored in SharePoint and that inside of Fabric. 0:59:13.928 –> 0:59:29.48 Joe Steiner It’s done. There’s a consumption model for that, which we can help you kind of understand how that would, how those charges would be incurred with that, the little bit of a calculator to that that that we can, we can help you walk through if you’d like. 0:59:29.928 –> 0:59:46.568 Joe Steiner I’ll see this. How does Fabric compare to other platforms and products like Databricks, Snowflake with Tableau and or Power BI? So if you’re not seeing wide scale adoption of Fabric, we we actually. 0:59:46.888 –> 1:0:2.968 Joe Steiner We’re seeing a lot of projects in that space. I think fabric’s been around for beginnings of it was probably 3 1/2 years ago and again, a lot of the underlying structures. 1:0:3.48 –> 1:0:18.288 Joe Steiner Behind fabric have been used. So those that had already built these things are just now some of them are starting to come over and say, OK, I’m going to change this over to more of a fabric style architecture. 1:0:18.568 –> 1:0:37.128 Joe Steiner I think you know over time here there is some functionality that you could do things more natively within the Azure data services space that have been brought into fabric over time. So the the offerings evolved, the tooling was always there, but they hadn’t folded it into fabric yet. 1:0:37.128 –> 1:0:56.368 Joe Steiner As far as things like Databricks, Snowflake, I would say there’s there’s at least, you know, with like a Snowflake, there’s probably 90% there. I can still leverage Databricks in here, actually if you have Snowflake. 1:0:56.968 –> 1:1:16.408 Joe Steiner There are ways to tie that into here at some levels too, although you start to get some redundancies there. But Databricks is something that there I can have some ties in there. I think Fabric is building more and more redundancies with what Databricks was offering the customers. 1:1:16.808 –> 1:1:34.728 Joe Steiner Previously. So again, it’s kind of an evolving platform. It really depends on what your needs were within Databricks. If you’re using the higher end functionality that Databricks and or Snowflake can provide, there may still be things that Fabric can’t do there. 1:1:35.128 –> 1:1:50.848 Joe Steiner But they’ve they’ve certainly increased its capabilities dramatically over the last few years. So we are actually having we’ve we’ve we’re doing a number of Fabric projects here so but you know it. 1:1:51.88 –> 1:2:9.888 Joe Steiner Yeah, I’m not going to say it’s the full functionality of a data bricks or Snowflake, but for some customers it’s more than enough. And so it kind of depends on what your data needs are there I guess. So you know we we are, we are actually seeing some significant growth in in fabric over the last year I would say. 1:2:10.168 –> 1:2:11.248 Joe Steiner That’s all for what that’s worth. 1:2:14.608 –> 1:2:14.768 Joe Steiner OK. 1:2:18.648 –> 1:2:20.128 Joe Steiner Uh, I think. 1:2:21.368 –> 1:2:41.88 Joe Steiner Those were the questions that I had. So again, we are happy to talk to you after this too. If anyone wants to reach out, happy to just have a conversation and we can, we can drill through some more of this. But I think we’re about at the end of our time here today and just hope everybody has a great day and we look forward to talking to you again in the future. 1:2:41.888 –> 1:2:42.368 Joe Steiner Take care.