Conference / Talks / Hasura in Enterprise

Hasura in Enterprise





The last thing I'm going to spend a few minutes on is just talking about some of these use cases that we have seen amongst our enterprise users and taking you folks over that in the few minutes that I have left.



First, let's start off with application modernization. This is a pretty common use case. When you think about application modernization, usually what's happening is that you have an application, it exists, you want to modernize it, which means that you want to move it to the cloud, typically, or make it more cloud native than it is. And you want to move it to a newer technology stack. Usually driven by the fact that that's kind of where developers skill sets are, and also where you can build better experiences for your end user, That's kind of driving that modernization.



But as I'm sure, all of us are intimately familiar with, any rewrite is always behind schedule and always more expensive than you thought it would be. Because it's a rewrite. When you kind of think about what needs to be done in this modernization step, you... Imagine it's a full blown modernization, you need to move databases, you need to move APIs, you need to move the app. All three layers need to move. So when you move the data to new database, you have two options. Maybe you do one time migration or maybe you do dual write. You keep the older system up to date and you keep bringing the new system up to date as well. That's kind of a dual write that you have when you're in that process of moving to a new database.



That's [inaudible 00:01:35] that you set up. Second is, you rewrite the APIs into a new way of those using those APIs. And third, you rewrite the consumer of those APIs typically in application. In some older applications, these two layers are combined into one app. There was no API, it was just an application, like a Java application on the desktop or something. With Hasura, step two is the step that gets reduced and that has had a massive amount of impact where these projects would've taken a year have now just taken a quarter. Because the work here is this tooling to be able to kind of do that in a way, and this work needs to be done anyway. You're primary rewriting that application for speed. So you, you have a good stack, you're kind of able to build that quickly.



This piece in the middle is a pretty big time suck and that gets entirely replaced or a big part of it gets replaced. And then some powerful benefits that we've seen is that it makes the whole rewrite process very fast, but it also de-risks some of that entire migration process. Instead of thinking about, "Oh, we're going to have to do this migration. It's going to take 10 years. It's going to take two years. Now it becomes trackable. But now it's like, "Hey, actually I think we might be able to get this done this quarter." You can de-risk that kind migration process because of that condensed timeline or because of attractable timeline. There's, obviously because of using Hasura's improved performance SLAs and improved security on the pieces that Hasura is taking care of, which ends up being a big chunk of the application usually and that has a massive amount of impact as well.



People are like, "Hey, wait, I have not seen p9 commands like this before. That's amazing." Which is nice. A variation of that and that's something that's going to start happening more than what we started seeing signs of today already, is database migration. Which is awesome. Especially as starting to support multiple databases. On the left, you have an old stack. What you want to do is you want to move that data to a new database. One of the big challenges there is that while you're doing this migration, you want a stable API across the old system and the new system. Building and maintaining a stable API that is going to work across two systems is complicated. It's hard. Hasura automates that process for a large part of it, that can be automated.



That makes this database migration modernization tractable. Effectively, there's two things in that that are critical to making this database migration story successful. One, is of course, native support for multiple databases. Which means that you have this ability to migrate or set up migrations or set up a stable API, essentially, that returns the same shape of stuff across these databases. As we add support from more stuff, it'll kind of increase. The second piece that's also really important here is [inaudible 00:04:29] draft endpoints, because that allows us to not think of our API as this powerful API that we want, but it allows us to think of this as restricted API that we want to provide and we can keep changing the implementation of this API when we move from one system to another or one database to another. Those two pieces become really important.



The next piece I'll quickly talk about and is the last thing that I have, I think I'm almost out of time, is that we can replace direct database access as well. That's kind of a use case that we're seeing. One variation of the at is, your data is on-prem, your workloads are on the cloud or their cloud native, that workload can't access on-prem database, because same problems, TCP, you want to guarantee quality of service, you want security. Typically, you have to build that layer in the middle as an API. Now you don't have to because you can just stick Hasura in. Another variation of this is that the data already exists, other teams want it, and so far people did not have an API for teams to access their data. Either people have to directly connect to the database, which nobody wants to do.



No team that owns a database, wants to give another team direct access to the database. It's just not a thing. Nobody likes it. It's not cool. It's my data stay away from my data. Unless I can give it to you from an API, and then I can control security on it. And then it's fine. Then I'll be like, "Yeah, yeah, yeah, whatever. I'm going to make sure that you can access only this portion." Because it's an API, I'll have some API tooling or I'll use Hasura to enforce certain quality of service things. In this case, when that team needs to expose data, Hasura becomes again, very useful in saying that you don't have to manually send data back and forth. You don't have to build that API. You can API file that data. Those are kind of the two variations of use cases and impactful use cases that we see with Hasura Enterprise users when they think about replacing direct database access.



The last, which is with my main, is that you cannot just build an enterprise app. And sometimes when you are building apps rapidly in enterprise, you won't hear this phrase too often, "build apps rapidly in enterprise" because it doesn't happen. The reason why it doesn't happen is not because nobody wants to do it, or people are bad, it's just because it's complicated because adhering to Enterprise grade SLAs is hard. There's compliance, there's security, there's performance, there's authentication, [inaudible 00:06:52] integration. It's painful and it's not easy. You can't just go in and use an entirely new stack as well. You can't go in and say, "Hey, I'm going to change everything here." You need some level of predictability and gradually maybe introduce new technologies in the new project that you have.



And, of course, there is one of the use cases that most of us are very familiar with is that it's pretty easy to set up on a new application quickly with Hasura. That works in enterprise as well and we see kind of a massive impact, especially with innovation teams in enterprise.



That's me a little bit over time but hopefully this painted a little bit of a picture in how we think about Hasura, how our roadmap kind of aligns with a bunch of the things that we're doing, and also some ways use cases and patterns of thinking when people have been successful in their usage with Hasura. I'll take some questions on the track but I'm out of time. Thank you for listening.

End of transcript


We'll go over the major use cases that Hasura is being used for: rapid execution on greenfield projects, application modernization initiatives as well as replacing direct database access with data APIs. Tanmai will take us over how Hasura is being adopted in enterprise incrementally, share the benefits experienced by users in enterprise as well as take us over commercial features that are offered to ensure you can successfully run Hasura in production in enterprise.

© 2024 Hasura Inc. All rights reserved