CockroachDB Update – Nate Stewart, Cockroach Labs

This is texturing TV. Hey everyone, welcome back to techstrung TV Our Guest is actually he's a returning guest for texturing. I think he hasn't been on though with us and maybe a year.

Let me introduce you to Nate Stewart from cockroach TV. Hey, Nate, welcome back. Ellen thanks so much for having me on your show.

It's a great. It's a great thing to have you on. I'm sorry.

It's been so long since you were on it's not like cockroach TV hasn't been doing anything. But you know, that's these have these schedules work. But in any event, hey, man welcome.

Nate before we get into kind of the news of the day. I wanted to give our listeners viewers a little bit of background about you and the company, so if you don't mind, let's hear a little bit about Nate Stewart. Yes, sir.

So, I'm Nate Stewart. I'm the chief product officer at cockroach Labs. So what we do at cockroach Labs is we make a distributed SQL database that makes it easy to build applications and services that are resilient that are scalable and extremely low latency and we work with some of the largest and fastest growing companies in the world to really help them streamline their modernization in their new app development initiatives.

Yep, and so Nate. I I kind of know a little bit about the Cockroach story. But let me let me play shell here for a bit.

So in in a world where we hear about new databases or new types of databases pretty frequently time series databases graphql databases and you know, no no sequel sequel. How would you where does cockroach fit into that kind of framework? Yeah.

So we're a relational databases a sequel database and you know people databases have been around since the the 70s and we're actually in a really interesting point in time because initially relational databases, they always ran on a single like very big machine, right? If you wanted to get more customers running on your platform you had to get a larger and larger machine and eventually you run out of machine and things get very very expensive very fast. And so what happened, you know about a decade ago is you start See the nosql movement where people are saying hey we have to scale.

There's too much data data is exploding and we have to get out of this the single Machine World and start to spread things out the challenge with the the nosql approach was that you had to give something up to scale right? So it couldn't figure out how to keep transactions how to keep SQL and also get the scalability and the resilience and so that's why no Sequel and that, you know pros and cons there there's pros and cons with everything. But what was the Breakthrough for us is that because computer science continues to evolve because people realize that it's very helpful to developers to be able to have sequel where you can write Less in your application here and rely more on the database.

Keep we're still trying to figure out can we do both can we scale out while also having transactions? Well also having Sequel and that's what you get with distributed sequel. So Coach DB is the result of some breakthroughs in computer science that really sidestep the trade-off between scalability and SQL.

I love it. so it's kind of like having your cake and eating it too right where you still get your SQL, but you get your scalability and Yeah, I don't know if the audience is figured out but I know we've covered this in time pass cockroach TV. Of course got its name from it's survivability, right?

They always say even after a nuclear Holocaust only the Cockroaches. Well exactly we'll survive and you know, it's a great take on that. So before we jump into we're gonna talk about a new release here, but for people want to get more information on cockroach and go check it out.

What the URL. com and and then the relationship between cockroach TV and Cockroach labs Cockroach Labs is the company cockroach. DB is one of our products.

Got it. I just want to just lay in the groundwork here. And absolutely all right, let's jump into today's news.

Sure. So this is one of my favorite releases ever because of how it came together. So we're finally at the place where we've moved beyond having dozens of customers to hundreds of customers and what that means for a database company is we get more feedback, but we also get more opportunities to ruggedize and strengthen the core platform because people are using it in many different ways.

2 and there are three main elements to this release. The first is all about developer productivity and a world where everyone's trying to find more efficiencies. How can you get the same engineering team to be twice as productive three times as productive?

And so what we're trying to do is how can we improve the SQL capabilities to again give developers more time back. In this case. We took one of the most popular features of postgres, which is called user defined functions where you can run code directly in the database rather than having to Rewrite it and manage it at the application tier again.

It gives developers more time back. The other exciting update is our approach to building real-time applications. So you can imagine you have your your database that keeps all of your customer records and think what you want to run analytical queries.

How much money did I make in the last month? How much money did I make today? Traditionally you'd have to do that in a one-off process.

I'll run the report every night at 1:00 am now with cockroach TB. We can stream those updates to your analytical system and they can run in real time. This is very powerful for customers like doordash, you know, they have their their Dash smart, you know convenience store platform and they use this capability to do real-time inventory management.

Could you can imagine you go on your mobile app, you're trying to buy some potato chips you put in your cart you check out and then it's not there no one likes that and so with cock which CBS real support real-time application now doordash can do real time Inventory management. And in this release we That significantly more efficient again getting more with less. And then the last big part of this release is all about operator experience and how can we help people run these increasingly large distributed clusters in production.

And so we launched this capability called intelligent insights because from the point of view of the developer, it looks very much like a postgres database but again behind the scenes we have many machines moving data all around the world. And so when things may slow down or if things are more expensive than you would think you need a different playbook for getting back into a good spot. And so rather than relying on Professional Services or documentation, you know, those are both great for cockroach.

We're trying to figure out how can we help the database give you hints on how to fix things specific example. Johanna Labs is one of our great customers there and that iot space. It's been out of Panasonic and they had a they had to me in is your indexes which makes queries fast and what that means in practice is that their database was too expensive and so now cockroach should be could say Hey, you have three unused indices and this is costing you x amount of dollars.

Maybe you want to remove one of those and that's the type of capability that we've built into the the product itself. So those are the three main updates from the latest release. Absolutely.

Hey, you know we jumped right into it and you were so excited about these features. I didn't want to stop you, but I I need to kind of frame a little bit for the audience. What what is this latest release?

What what's it called? 2. It's it's not so in your naming.

Convention is this considered a major release or it's just a point release. This is a major release so our name is invention the first Instead of digits is the the year and then the second set of digits is the number of major releases. So this is the second major relation major release for this year.

com, what do you go to cockroach DB for this one? com and then you can either download the binary or what's the most exciting get started in our Cloud product? Because again distributed systems lots of machines we wrap all of our releases in this broader Cloud platform.

So you can run data intensive workloads at a fraction of the cost using our management infrastructure. So again, just less operations really focus. Yeah.

BB is a service baby so Nate, this was released if I'm not mistaken December 6th. Yes, and so it's been out a couple weeks. What kind of feedback you've been hearing from people.

Yeah, the the user defined functions in particular was one of our our most requested features and it's one that we held off on and building for for some time because initially with that cockroach CB people were using it to build their next Generation applications. And for those people were relying more on microservices and you know, trying to spread things out and Federate things and what happened was once people saw that cockroach to be actually did deliver on distributed sequel. That was the hot moment for a lot of people and they're like, hey, we have all these postgres workloads.

Let's just migrate them on the Cockroach and one of the barriers was the support for you to find functions. And so with the release of that that has unlocked tremendous potential for people to modernize even more of their application tier and what's really exciting and their data here as well. But what's really exciting about the our approach to use it to find functions versus what you see with say postgres is we are building it from the ground up to run and distributed environments.

So now it's not just going to be user defined functions on one machine. It's going to become distributed data functions where you can have these functions running close to data spread around the world close to your users all running on that same scared cockroachdb infrastructure. It's cool.

Yeah, absolutely. I mean and you think as we taking this full circle back to the beginning of our conversation, we're doing this on a kind of I don't want to use the word infinitely scalable, but highly highly scalable. Sequel-based server because one of the things Nate that I've been seeing, you know, I have a good seat right here.

I get to talk to a bunch of people. One of the things I'm seeing is even these, you know, those newfangled tiller databases like graph or time seek or time series or what have you a lot of them. You know perform functionality and all this over here, but they they have sort of sequel back ends.

If you will right where they they work on Sequel and so You know. It to have a cockroach as that sequel could only help them with their scalability because a lot of them don't work unless you have really big. Data sets to begin with right.

I mean that that's where their value comes in. So it's an interesting company, you know relationship. Yeah, it's interesting.

We we actually see new databases that are being developed that are being built on cockroach DB Starburst. For example, this is a analytical database as a service. They use cockroach to be under the hood to manage a lot of their metadata.

Make sure that their database is always available. So we are saying that the foundation that we're providing is sparring the next wave of innovation in the database space. Absolutely.

Hey Nate, we're about out of time. But I just want to remind people again. 2.

Yes, and you can get it right on cockroach slabs. You can load down download the binary and set it up yourself. But you know if you want to just get up and running use the SAS version and it's there and already to rock and roll.

Exactly. We're seeing tremendous customer attraction. We've used that to improve the product that you can experience it all on our Cloud platform or download the binary.

It's great seeing you. You know, it's I now I understand why I saw you about a year ago. 1 or something.

2. Let's not wait till it's twenty three point one. Keep us back on what's happening with cockroach.

Okay, man. Okay. Thanks so much Allen.

2. We're gonna take a break on text drunk TV, and we'll be right back.