Skip to main content

Partner Summit Breakout: Migrating to Databricks with Lakebridge

Summary

  • Modernization imperative: Legacy data warehouses represent expensive technical debt; migration to Databricks enables leverage of enterprise data for higher-value analytics and AI.
  • Unified migration tooling: Lakebridge accelerates database and EDW migrations to Databricks, providing comprehensive coverage and cost-effective modernization for enterprise data estates.
  • Partner enablement strategy: Access Lakebridge through partners and the Databricks ecosystem, gaining field engineering assurance and cost-savings models for smooth transitions.

Partner Summit Breakout: Migrating to Databricks with Lakebridge

Watch: Partner Summit Breakout: Migrating to Databricks with Lakebridge
Talk by Guenia Izquierdo Delgado, Databricks Labs Manager, Databricks ; John Young, Managing Director - Field Engineering, Databricks
In February of this year, we announced on our acquisition of Bladebridge. We'll share the latest on this tool and how it can help our partners speed up migrations.

Chapters

FAQs

Why should organizations migrate from legacy data warehouses?

Legacy data warehouses are expensive anchors; migration unlocks data for modern analytics, AI, and higher-value use cases while reducing operational overhead and technical debt.

What is Lakebridge and how does it help with migrations?

Lakebridge, from the Bladebridge acquisition, is Databricks' migration tooling that streamlines the transition from EDW and on-premises databases to the Databricks Data and AI platform.

What coverage metrics demonstrate Lakebridge effectiveness?

The video shares migration landscape data and coverage metrics showing Lakebridge's ability to handle diverse data warehouse and database environments at scale.

How does Databricks support partners implementing migrations?

The video highlights professional services assurance, cost-savings models, and access through the partner ecosystem to ensure successful migrations for customers.

Full transcript

0:00 Introduction to Lakebridge

We are on. All right. Um, welcome. Welcome everyone. Hello. Thanks. Thanks. That's why the those eyes were peering up at us. I wonder what he's going to say. Um so uh welcome everyone uh to this special edition of migrations uh to data bricks with lake bridge. Why is it a special edition? Two reasons. Uh number one is we're seeing a flood and you probably are seeing this too a flood of interest and activity um around uh hey I want to explore or understand better EDW migrations or migrations from on-prem databases to data bricks.
Number two is that we um I think uh are at a an inflection and a kind of a shift in strategy where we're really baking and bringing the tooling together. And so I'm joined um with uh my esteemed colleague is Escardo Scardo. So close. so close. Um, who is the product engineering manager for our tooling u, our migration tooling. So, thank you. Welcome. Thank you. And with me, we have John Young, our MD for field engineering partners.
All right. So, uh, right, no surprise you're going to hear from me, uh, that legacy, uh, data warehouses are an expensive anchor for every enterprise and it's time to move, right? So, um, does anyone in here disagree with that? So, I just want to kind of sort of level set and feel free to say, yeah, I I I just I disagree with that. Okay, so on is not an expensive anchor. As long as you are just sure. Yeah. Yeah. Fair.

1:37 The case for modernization

Right. Uh so like copies of data isn't super effective is I think is the net of that. Right. Like hey why are we just going to continue to copy? let's figure out how to either migrate, federate, etc. Right? So, we're going to talk about a lot of that today. So, this one, this slide I I thought was kind of interesting, right? And um I think it speaks to not just legacy, but also to what I think is um challenging about migrations, right?
And we see this, which is this kind of mounting of uh legacy or technical debt, right? um which is built on a database, a data warehouse built on a database. Uh kind of challenging when we talk about like leveraging the data that's in there to other uh other use cases, higher value, more an you know analytics um merging with non uh data warehouse data etc. raw data where is that in the in the database? Um and then the obviously on the other side is the boxes, right, which um you know that's more traditionally sort of where we see that legacy concept.
Um they're expensive. Um there's no I and then you you get sort of you're locked in and then you're you're sort of um in this cycle of uh renewals and um and constant negotiation. So um I think we have something better. Definitely we do. Um if we contrast that with data bricks SQL um with data brick SQL we have worldclass performance and TCO because enterprises want to know that what they're getting is performant. They want to know that they get a they get value in their new system testing.

3:46 Databricks SQL platform

Um, they also want to make sure that uh when they migrate to a modern platform, their code is portable. It's it's open and it's ready to be uh integrated to the rest of their ecosystem. And last but not least, data brick SQL comes with batteries included because we have dashboards with AIBI which allow you to enable your customer and your customers customers to do self-s serve Q&A. So it comes as a no surprise that over 11,000 of our customers choose to use DBSQL as their data warehouse.
Well, let me tell you, I used to work for one of these uh in one of these slides in one in this slide. I'm gonna tell you which one. Um, and we struggled quite a bit because we didn't we didn't have DVDs at the time. So, we tried to make our own platform and try to make it as useful as possible, but the UX wasn't great. Um, so our ask of you today is that you you tell your customers. You are in front of the customers all the time.
We aren't. So we ask you to tell your customers about DBSQL. Tell them about the benefits of DBSQL, how it will make their lives easier and cheaper. And a quick show of hands, who is familiar with DBSQL? Awesome. Everyone's familiar with DBSQL. That's that is awesome. Um because it's not that old of a product from a cloud data warehouse perspective, right? It's only a couple of years old and uh we've seen since uh since its inception, right, continued improvements uh continued expansion of features.
Um which is why we um get rated very highly in um this uh industry, Gartner's industry. You've probably all seen this. Um I'm not sure who kind of uses these um magic quadrants. um but they they seem to carry a lot of weight and a lot of value right so we're a leader in database management um in um you know both forester and in there we go and uh just published recently I think it was two weeks ago or within the last couple of weeks um we are now defined as the leader on uh data science and ML which we've you know as as a a company that's was born in in data science and ML that's sort of to be expected But um it's pretty cool to be kind of out on the edge and be recognized as as as the leader um compared to some pretty um important um u company there.
Now uh it's not just uh machine learning. It's all-encompassing of AI and the things associated with that. So where uh and so what right does uh so we've we've got the rating and we've got the the recognition but and we've got a product but is it actually being used on on complex uh data states that you guys are all seeing? Definitely. And you can see so there are three very large customers up here. Um where again their data estate is huge.

7:01 Customer migration success

Um they had multiple data warehouses at the core of their business. Thousands of people quering these data warehouses every day. So when you're at a bit when you're with a customer like that and you're advising them on how to move forward on their data strategy, how do you make sure that they are democratizing their data, that they're simplifying their usage and they're making their process more streamlined? Well, if you look at these customers, you see that they not only had pretty sizable benefits from a cost and time perspective, but they also landed in a in an open format and an open uh platform.
So, it will really help them going forward. So, that is a winning situation. That is a winning formula for your customers. So, I want to add one little thing here uh impromptu totally ad hoc. So, I had lunch. I was sitting with lunch with um the person who manages datab bricks is datab bricks's um yeah the apostrophe datab bricks's um internal lakehouse and uh so we got to talking and in fact Sergey is presenting there like right now he's uh in the next room talking about data bricks on data bricks data bricks usage of data bricks and um So 40,000 uh it ranges from 40,000 tables to 100,000 tables in our internal um in our internal lakehouse.
And uh one of the things that he talked about so I I was like kind of pressing him like hey what what's really good what's not good and you know the thing that thing that he talks about um as being kind of differentiated is this ability around monitoring and quality of those tables and I think at scale that's the piece that I think when it comes together and what Gennine is talking about here with these large customers 90,000 users uh that's a lot of tables um that's sitting at or satin AT&T and GM.
So um I think uh you know you are at customers where there are very large on-rem data states uh DBSQL can handle it and what we're going to talk about from a tooling standpoint uh can handle that. So uh don't shy away from 100,000 tables or half a million tables definitely. So let's talk about how you do that. So we know that there are still a ton of legacy data warehouses and sources out there in the market and in fact the Amazon CEO says that um the global IT spend is it accounts for um 85% of it accounts for on-prem service uh systems.

10:02 Migration landscape data

So if you combine that with the expected growth of data warehousing to 70 billion in the next couple of years, well listen, we need to make migrations easier because there's going to be a lot of migrations happening. So uh how many uh are those migrations are going to happen? Well, I think it's actually going to depend on how many of you uh sort of combine or jump in and uh like get smart on the tooling, right? So we have uh right now is going to talk about this product, right?
We're going to come out and you're going to see um we've got a lot of source systems and more are being supported. Right now it's 10 plus. Um there are uh over 200 partners with expertise on the existing tooling that we've made available through the acquisition of Blade. And so many of you in the room have either used Bladebridge before um used some of our open source tools in VBA and Beta Bricks Labs or got your own scripts.
Um 2,000 so far of you have kind of signed up and taken and completed our our training at some level. So, and that's either on the analyzer or converter, but we're adding a bunch of new stuff to this, which I think is why you're all here to sort of hear a little bit about what that is. And um you know, it's it's um yeah. All right, next slide. All right, so one and two, right? So, first stop um you know, how do we do this?

11:51 Migration strategy formula

Why do we do this? Uh align with a partner, right? You've heard this. And if you in this morning's session with, we're partner first, right? So from a migrations perspective, we want to align with the partner first early in the process so that you're available to either execute on that profiling and analyzing of the data estate or we're assisting and providing that um already done to you. That's like the strategy align with a partner.
And then the number two piece is define your payback, right? We we used to be able to say, oh, the ROI could be um you know, more than a year or it could be two years or it could be whatever x number of months out from um you know, from when that migration takes place to when the customer is really seeing ROI. I think that window is uh shortening pretty dramatically uh so that we you know you know sort of benchmarking at a year right so it's costneutral effectively in a in a budget year um we're we're actually seeing that sort of shrink um some of that is is the cloud funding right and the interest in cloud funding and as as we mentioned earlier today we've got a lot of funding programs adjacent to cloud funding that you can sort of stack up and help drive um and cover many of the much of the services costs that you've you know that that are required to make this migration work.
Um but we're also seeing tooling and automation really shrink those timelines down. And so because of that, we see the ROI um happening much faster in less than 12 months for many with the exception of the very very large um estates um that's shrinking and customers are kind of expecting that. So we see a lot of different ways to do that. What else do we need to do? We need to scope our migration, right? And that's what I'm here to talk to you about.
But before we get there, um I want to let you in in a little secret um that you can't tell anybody, but um when we announced the acquisition of Blade um at the beginning of the year, that blog post where we announced that is one of our most read blog posts. Can't say that. Um out there that we've put out recently. So, and why is that you may ask? Um well, I think it's fairly obvious. Our customers do want to migrate to data bricks.

14:05 Cost savings model

They're interested in coming in. They're interested in learning how can they how can they do that but be effective um cost effective when they do that. And the best way that we have to make that cost effective is improving our tooling. Should we talk about Unity catalog? Yes, please. All right. So, uh everyone know what Unity catalog is? Another show of hands. All right. So, there's a few people in here that are like, "Hey, I've not heard of Unity."

14:50 Unity Catalog governance

But um I I do want to talk about unity u because it's super super important. Um the partner summit this morning there were a couple questions uh from you know from the audience that sort of let the feeling like hey not everybody really knows what Unity and what Unity does what Unity offers and the extent of it. It's not just simply a catalog. It's a it's a security and governance framework. It covers lineage. It does data.
It does models. So you have the entire data estate whether that estate is actually stored and running within the lakehouse or if it's federated and you're bringing that data in through some other means. Uh Arcelon mentioned there's a number of announcements coming out that uh I think I I personally am excited about. I I have a data science background and I used to hate sort of building pipelines. I just wanted to grab the data and then build the model and um this some things that are coming out are going to make that a lot simpler.
So I don't have to call a data engineer. Uh I can actually just do that by itself. um federation right uh lakeflow uh connect you're going to be able to connect you can right now to uh stores like uh SAS apps like uh Salesforce Service Now workday um some of the foundations of SQL Server and there's a lot more kind of coming into that uh pretty quickly uh including some of the on-prem uh including some of the on-prem sources so you sort of start to imagine a migration maybe like looks a little different, right?
It's not this sort of like flip the switch. Hey, we got to get everything running, test it all, and then we flip the switch and now everybody's on a new system. You can actually sort of move things and uh live in two worlds for a period of time, especially if you've got this um this sort of governance through Unity Catalog. So, super important. Um, you're going to hear about Unity catalog and the requirement to have partners really specialize in Unity Catalog and it's really specializing in security governance.
So, it's going to be Unity catalog combined with HADR combined with security frameworks um, and kind of what that looks like and some best practices around that within the within the platform. So, more to come. It's not just Unity, it's Unity plus some of these other features that um you know I think live in an on-prem world and will now be able to sort of also be visible in um in the cloud with DBS SQL. So speaking of migrations, um Garner put out this interesting uh image where they equate migrations DBMS migrations to heart surgery.

17:52 Current vs desired state

Um the problem is these these data warehouses are at the center of the operation of our enterprises. So when you remove that and you have to be very careful about not disrupting the system. So the reality is that in the current state um migrations are unreliable. Um it's difficult to estimate the cost, the effort, the time they will take. Um the reality is that many tools out there that only do partial translations. So you end up having to do a lot of manual effort to maybe they do only the SQL conversion and they leave all the ETL baggage behind that you have to do by hand after.
Um and that results that at the end you don't have a clear way of telling the business, hey listen, you can shut off that other system. We can prove it. We show you the the truth that this is all made safely here. So with automation tooling frameworks, we're really aiming to fix this. We we're hoping to automate all of it. So we want to make sure that you can pro present your customers with predictable effort and estimates on your migrations so that they don't run over, they don't go over time, they don't go over budget.
We want to present you with a tool that you can use to migrate a broadsource um set of systems of SQL and ETL. Um, and we also want to present you with tools that you can use to also talk to your customers and boost their confidence that yes, the migration you did is correct and it's done and they can switch it off because they can unplug it. So, at the beginning of this year, as I mentioned before, uh, we announced the acquisition of Blade Bridge um, and by combining their um, the tooling that they brought and our own internal tooling that we had created.

19:35 The Lakebridge solution

Today we want to talk to you about Lake Bridge. So, Lakebridge is a unified free solution for migrating more than 11 sources um from data warehouses and SQL systems into data brick SQL. Also, I want to mention that while we were building Lake Bridge, we realized that we're not a migration company, right? What we want to do is that we as data bricks, we want to empower you, our partners, to make sure that you have the tooling you need to migrate your customers to data bricks.
So we want to give you a full end to end set of tools to help you on that migration. So first up, we want to give you a profiler. What is a profiler? I get this question all the time. Everybody gets confused. So a profiler is you're going to connect it to the source system of your customer and you're going to understand what is their data state, how many tables they have, how many users do they have querying, what is their concurrency, what are the times that data warehouse is really used that you don't want to be messing around with things that do not disrupt their their their business, right?
So you want to plan your migration around those times too. Then you have an analyzer which is more understood by the business now uh which is of course looking at the source code and understanding what's the complexity that you're going to have to incur to migrate that code into data bricks. How difficult is this codebase then we have the conversion. This is a need whenever anybody talks about tooling they always talk about conversion.
There's so much more to it but anyways conversion really important. Um conversion is the one module where we really thought about you. So, Lakeidge comes with two converters out of the box ready for you to use, but we made the converter the converter module pluggable because we know that many of you have your own tooling for doing conversion, right? So, what we want to do is make sure that our tool is also friendly to yours.
So we we have defined a spec using I'm going to geek out but we're using uh a uh language server protocol uh to do this this interface with your um converters that you can choose to plug into language and what is the benefit you might ask. Right now there's not a lot. So the main one is that you get uh inline um syntax validation for the code converter. But in the future we plan to make all of these system share information between each other.
So you will have data from the profiler to the analyzer being shared to the plugged converters so that you can use that data to make better conversion options. Okay. And the last step is the validation is the reconciliation to make sure that it's providing the business users with the confidence to say yes I see that the data that I moved is truly what I got on the other side. Okay. So with Lakebridge, we expect that you will be able to automate 80% of your migration tasks and accelerate your migration projects by 50%.

22:39 Competitive landscape

So talk about competitors. Let's do let's talk about competitive landscape. Right. So some of you know this uh this better than um than I do. So I'm going to touch on these few points. We were nuts with the with the build. Okay. I forgot that this was an extensive build slide. Um so right like snow convert uh and snow park uh pipelines right um this is partial conversion. Um if anybody's used it you you know that it has a sort of pretty significant limit.
Um, you know, we we don't have that limit um on Lake Bridge and that's like the things and the features that are going to roll out on Lake Bridge are going to sort of work end to end. That's the way it's it's going to roll out. It's going to go all right, we get to do all the way through. We're not going to sort of piece meal it out and short circuit um some of your efforts. So, you're not, you know, sure of the risk that you're adopting.
Second on BigQuery, right? There's a bunch of tools in BigQuery. Uh that's almost really not a a a Google provided. It's just like scripts effectively. Um it's uh you know super fragmented. Um we want as as said to provide this single tool. Fabric has um a migration assistant. That migration assistance is fairly limited. Um certainly uh their focus to drive workloads or pull workloads off of Synapse. Uh maybe some of you have actually worked on that or worked with it.
Um you know, right? I mean, if you if you want to move off Synapse onto fabric, great. Um but if you want something else, uh then then that's limited. And Lake Bridge really kind of provides this broader set of capabilities to go to an open platform, an open lakehouse and then do that from multiple sources. Synapse will be one of them. Um BigQuery, uh Snowflake, etc. Uh but what if I need uh what if I need help, right? and um and some of you have worked on um worked on migrations with us and uh and are familiar with our our professional services organization.

25:03 Databricks PS assurance

Um I'm going to do it kind of real brief. Uh our professional services organization, the strategy is partner first, right? I I started with that to say we need to number one align with a partner. That's the strategy. That's the strategy with professional services. I actually sit in the professional services org. So, um, I work with who's who's who heads that and we are always driving, he's got metrics that are really about driving partner first and partner priming those migrations.
Um, when there is a stuck, when there is a need, when there's support needed, they do that through assurance model. And um, and that's the sort of preferred way to get help. It's through an assurance model. We deliver through onshore and offshore resources depending on how you do that or what your needs are. Um and there are relatively few exceptions and those exceptions have to be made at a pretty senior level to um move away from partner priming that um that migration.
So what I think uh what what what I've seen work um in migrations of large scale is when we bring uh you know when we bring you in early or when you're already there responding to an RFP um and are looking for assistance the the sooner we can lock up together um with our PS orgs who can say here's the best practices go for it um conquer and you know make the migration work or here are the resources. Oh you you know there's some complexity that's appearing out of the analyzer that you may need assistance with and configuration assistance with.
We'll provide that um as well but it then gets you get the opportunity to package that into the migration and we have funding through accelerate to support those kinds of things. I know I see many a couple of um faces in here where I know that um I've been personally involved in getting you accelerate dollars and getting your projects um that uh that support. So, uh PS uh PS for good is the way I sort of think about it.
Uh PS assurance for good. It's really to help customers move fast um get you best practices and there's a lot of assets um for you and uh if that's not enough then you know talk to me and we'll we'll figure out what we need to do to get more um to more more support for you. Um now, uh now that I know a what Lakebridge is, I know how to get support. Uh then, uh then what? Then what? Yeah. Um I hear these guys like numbers.
So let's talk about what do you get by using lridge, right? So what do you have right now? Again, we're building this as fast as we can, but what do we have right now that you can access right now? Um, on the profiler, we have our first profiler out for Synapse. The analyzer, I said 30 here, but really the analyzer, anything that queries data, guys, just throw it at it. It's really good. Um, on the conversion side, does that mean like a hundred?

28:01 Lakebridge coverage metrics

Maybe. Are there are there 100 things that can query data? Maybe. I don't know. Probably. Could be. Um, on the conversion side, right now we're sitting at around up to 70% of overall success with conversion rates. Um we have over 11 um sources both SQL and ETL that you can convert from 100% coverage on tables and views. We're working on store pros. We know storebooks are hard. Um but out of the box we will be able to convert to data brick store pros and SQL scripting.
Um on the validation side it's a really scalable and performant um reconciliation module. So you can use it at ease no matter how big the tables that you're reconciling are. So, one of the things that I'm excited about uh that Guinea was talking um the engineering team was working on is right stored proc to stored proc um you know okay but stored proc to DT or declarative pipeline uh much better much more efficient um lots more opportunity for optimization tuning and really uh quality checking you know those transformations so that is being worked worked on and if you've got insights, we want uh to hear about those and collaborate with you.
Um I hear uh I hear that there's another talk uh about Lake Bridge during the keynote where they might be talking about some other things on Thursday. On Thursday, that's right. Is that right? Okay. So, um so we should Yeah, they should encourage everyone to sort of pay attention on Thursday during Arcelon's keynote. All right. So um my first so then my next question is you mentioned free but um what's uh what you know what's what's the catch um what does it cost actually I don't I don't believe that guinea is there's there's there's no cost there's nothing on this slide um this is purposely left blank because there's no cost to using this is open source you can go you can check it out today right now and you'll be able to use it with your customers right away um the fact that it's openour ource it's also again as I mentioned before because we're keeping you in mind we want to work with you because again you do so many of these help us get contribution into get contributions into Lakebridge we would love to partner with our partners um to make this product better so now you know my name ping me let's work together all right uh so now that you know does that mean they're not required to get assurance services either no they're to configuration.

30:19 Lakebridge cost and access

Okay, so not required to get PS assurance as part of the migrations. However, as I mentioned earlier, sometimes you get into that spot where PSur is really fit for the customer. Um, and we want to make that really fast and easy for you to do and also identify vehicles for funding that. So it's not like, okay, what do I do now? I've got this extra cost that I don't know how to deal with. uh we're like we are truly like want you to be successful on these migrations.
We want it to go fast and we want you to shine in the in the customer's eyes and I know that cost is a big component of this. So um yeah. So what do we what what's our call to action or calls to action I guess? Yeah. So the first one mentioned make sure you're at the keynote on Thursday. I mean make sure you're at the keynote tomorrow too but on the Thursday one uh specifically uh for sure. Um the second one is that we're doing a deep dive into Lakebridge and all the details of the different modules today.

31:56 Partner call to action

I covered it very high level but if you're interested in learning more today there's a talk at 5:30. I wish I knew the room but I'm very confused with this building so I don't know. It's called comprehensive data warehouse migrations to DBSQL. Um and it's happening today at 5:30. That's the first one. That's the first one. All right. So, the second one, uh, I mentioned this sort of flood earlier of, uh, customers that are interested or seeming to be interested in, um, trying to figure out how to go, how do I get into the lakehouse.
So, what we would love for you to do is um, is to start to position these assessments uh, free, which we will support in that pursuit. So again, I mentioned the PS assurance. It's not a you don't have to pay to get the support, but we we want to engage early and support you in these assessments. So let's get in to your customer and say, "All right, uh let's let's do something together because data bricks said we can do this and um I and you know, you're working with them.
You're the adviser, the trusted adviser for this customer. uh we'd love you to sort of step into that and be bold and ask for um those free assessments or ask to take part and initiate that. Um we also uh to G's point u earlier around the collaboration right we know that uh a few partners in here that are building um agents um which I think effectively when you look at migration tooling it's really a patchwork of agents scripts and agents but effectively different things that are really good at very specific tasks right and Um, we think that some of the things that are coming out, uh, and I think we we're pretty pretty certain of this, um, that some of the things that are coming out tomorrow and Thursday, um, if you're not familiar with our agent framework, our agent bricks, as um, Arcelon mentioned earlier today, um, we want you to step into that and then start to collaborate, um, with your ideas around how to leverage the platform to add to that patchwork um, or to that um, that quote of migration tooling.
I think um what you get from a benefit there is that we want to do two things. One is um uh help you understand the platform better and by working on that in a collaborative way with some of our sort of agent experts if that is is a thing it probably is going to become a thing. Um that'll help you. Number two is uh we want to fund um in exchange for that collaboration we want to fund um through DB Express credits uh TBD and what that looks like and but it's going to be meaningful and um and that'll help you generate um more opportunity create more demos and other things that support your pursuits with customers.

34:43 Agent framework and credits

So, uh, it's kind of a little bit of a flywheel to get and, uh, to get build, collaborate, and get smart on the platform. All right. Question. Yeah. Microphones here. I can do the simple one. All right. You go. Shoot. Shout. Shoot. Shout. Oh, yeah. DV Express. Anyone know anyone familiar with Can I get that question again, please? Oh, you've got it. Okay. Oh, okay. All right. Yeah. I just wanted to Yeah. DB learn more about DB Express credits.

35:58 Tech summit registration

Sure. Uh DB Express. So, right, that's our serverless offering. Um you're going to hear a little bit more about what that uh what that looks like tomorrow, but many of you um you know already had this opportunity to sort of it's the SAS version of um of data bricks. DB Express is effectively the SAS version. That's the way I think of it. It's the SAS version of data bricks. I go in, click, push a button, boom, the entire sort of infrastructure is set up.
It's all serverless. It has all the features um of data bricks and it allows for a rapid prototyping. Uh partners use it for demo, for testing. They do some experimentation on it. And in um in in the partner program, we're rolling out Sonel, who may have be be presenting in in the same um same time zone here. um she's uh providing the opportunity for you as a partner to say, "Hey, I've got this customer who's never worked with data bricks before and we want a simple fast path to turning on data bricks for them and getting data in the we're going to actually provide you with credits on DV Express for you to stand that up."
So, super cool. And then the exchange here of hey, let's work on tooling together is really to add to your pile of um of credits that you that you might already sort of be given. Let's wrap up with the call to actions first and then take more questions. Oh, sure. Yeah. Uh okay. Two things. All right. Two things. Number one, um take your phones out, click on this QR code. Uh oh, we have uh we have text over the QR code.

37:41 Q&A and closing remarks

I bet you it's not going to work. I just noticed that. Um, if we stop talking, maybe. Does it work? It works. No. Oh, that's amazing. Um, so TechSummit. Who's been to TechSummit? All right. Yes. Yes. Yes. All right. So, uh, the end of this, I'd love to see everyone at TechSummit. So, when I come back next year and ask for a raise of hands, I want to see everybody. TechSummit is the one opportunity for you to sit with our Fieldge teams who are also going through the product and learning about the product and the platform.
So, they're getting learning and you're getting the same learning at the same time and it's an exclusive offering to partners. Um, there's a couple of ways to get into this. The next one is at the end of August, early September. So, please opt in because we can't just uh spam you with emails. You have to opt in and say, "Hey, I'm interested. Send me some info about it." Um, it it's really good. Uh, it's like three days and it's deep dive.
Uh, it is like top priority from a learning standpoint within FE. So, super important. And I also wanted to leave you with this um for the documentation, the GitHub address for the repo. How can you find it in Pi? Um so if you scan that, that'll take you to the documentation. Um and you'll be able to learn more about LakeBridge. And this is cool documentation. I went out there yesterday seeing it for the first time and I was like, oh yeah.
Uh compared to what we used to see, Lake Bridge is like, whoa, it's blowing my mind. Um, and it's just starting, right? So, Thursday, you're going to hear a little bit more. Um, but fun stuff. Stop in at 5:30. I I don't know if we're going to get pushed out, but if you have some questions right now, we could probably uh Oh, she's smiling saying I'm definitely not going to get pushed out. Well, we might be. Not at all.
Thank you. Yeah. That you're talking about the historical lineage that you also want to carry forward. Uh yeah. Yeah. Uh so good question. question was really um about how do we right it's one thing moving the data and the queries and all of that but it's the but the next question is what about lineage and what about all of the history that um especially in regulated we want to maintain we don't want to lose how do we do that I think um you know sort of the short answer is uh stay tuned because we have um some thoughts on how we accomplish that um it's It's not available yet, but it's on the road.
It's it's Yes, it's it's in the queue. It's in the road map. Yeah, I'll be really quick. Uh hi. Um what I wanted to know is I've worked with some of the federation before on like Oracle for instance, and my understanding is the performance of each federation varies a lot. Can you kind of give me an idea of how it performs on Postgress, how it performs on MSSQL, etc. if you know? Yeah, I don't have those. Uh, so you're going to see a bunch of stuff coming out on performance um across the board.
So, what one thing that I one thing that I'll just share, right, is um what I've seen is some are choosing federation when we probably would look at and say CDC is a better option, right? And you're going to get more performance instead of fed instead of a pure federation. or if I want to hold on a federation, then I better do a full copy. So, LakeFlow is intended to sort of get you into a little bit better place there, but we continue to work on improvements on performance of federation and query federation.
And you're right, um, each platform is a little bit different or each source is a little bit different that we're that we're coming from. Um but uh more more to come on performance of federation federated tables in the next couple days. Sorry, one last quick question. Uh I understand that blade bridge is and lake bridge are not the same thing. br blade bridge is a plug-in to lakebridge. So the links that you shared that open source that for the core lake like just the original lake bridge or does that include that blade bridge ETL conversion and if not then what is the actual way for the partners to access the blade bridge converter today?
How do we actually access it if it's not on GitHub or is it? So the code is not available on GitHub for the blade bridge converter. What you will see is the open protocol calling the lakebridge converter. the blade brbridge converter. So you can use it from the lakebridge API. So when you do lake brbridge transf to execute the code conversion you can specify okay use the blade brbridge converter to do that conversion or use the other converter that is available and we've provided you with a matrix of which one to use when we're fixing.
Okay. So long story short, if we are migrating from Informatica, let's say to databicks today, then those links that you shared on GitHub plus the pages with the documentations are enough and sufficient for us to begin that conversion right away. Is that correct? 100%. And we have documentation in there. If you want to extend the configurations to make the conversions better, that documentation is also there. Um, so yeah, it's okay.
It's okay. You're welcome. All right. Thanks everyone. Thank you.

Learn more about the Databricks Data and AI platform.

The information provided herein is for general informational purposes only and may not reflect the most current product capabilities or configurations.