text stringlengths 0 2.68k |
|---|
**Sugu Sougoumarane:** Yes, yes. So there is a philosophy in -- so we had that luxury, and that's the reason... I think that's the main reason why Vitess succeeded where others have failed, is because nothing that would not scale could be even written in Vitess, because the next day YouTube would go down. |
**Jerod Santo:** Right. |
**Sugu Sougoumarane:** We have created a few outages due to those things. So if something was not going to scale, we would know right away. So we had that luxury. So everything that went into Vitess had to work at scale. So that was one thing. And the advantage we have is we have those learnings. So we know what not to... |
**Jerod Santo:** Sure. But don't you want to test it as you go, and make sure that you're doing it right? |
**Sugu Sougoumarane:** On the testing side, what we over time developed - that is something we we had the luxury of not having to do in the early days at YouTube - is we had to develop really, really, really, really thorough tests. The reason was because at some point of time we would be making changes in the Vitess so... |
**Jerod Santo:** Could you actually bring the test suite over, or at least parts of it, from Vitess? I mean, because that's a pretty robust test suite. If you could bring if you could port that somehow relatively easily... |
**Sugu Sougoumarane:** Yeah. Anything we can copy from Vitess, we are going to. Definitely, yeah. There are some parts which we'll just copy. There's no reason to change anything there. So those parts we'll just copy, yeah. |
**Adam Stacoviak:** Have you looked at deterministic simulation testing, similar to what Turso is doing? And what was the other database, Jerod, that you had \[unintelligible 01:01:27.11\] |
**Jerod Santo:** TigerBeetle. |
**Adam Stacoviak:** TigerBeetle, yeah. Have you looked into that, and does that translate? Or is the test base -- |
**Sugu Sougoumarane:** I need to look this up. But in Vitess we probably have the opposite, which is called the fuzzer. Is that the same thing? |
**Jerod Santo:** Similar. Yeah, similar. |
**Sugu Sougoumarane:** \[01:01:46.24\] So we do run the fuzzer, and the fuzzer does not miss -- as far as I know, it misses nothing. I wasn't there when they implemented it at PlanetScale. It was someone called \[unintelligible 01:02:00.11\] who did that test suite. He said that the fuzzer was so good that it found bug... |
**Adam Stacoviak:** That's good. |
**Jerod Santo:** That's a pretty good fuzzer. |
**Sugu Sougoumarane:** So I have pretty high confidence in that. But yeah, whatever it takes. |
**Adam Stacoviak:** Well, we have a couple of past shows we can recommend for you as well that might give you some insights to -- just some of the behind the scenes; not exactly that how to implement it, but more fodder for your request. |
**Sugu Sougoumarane:** Yeah, yeah, definitely. |
**Jerod Santo:** Well, it's all very exciting. I assume the licensing is going to be similar to Vitess, or similar to what Supabase normally does. Is there any sort of gotchas in there, in terms of the open source side of it? |
**Sugu Sougoumarane:** No, no. Apache is what we're planning to do. Vitess was actually BSD once upon a time, until we joined the CNCF, and we changed it to Apache. But yeah, it's going to be Apache. |
**Jerod Santo:** Does Multigres make sense in the CNCF as well, or no? |
**Sugu Sougoumarane:** It's an option. It's definitely an option. We even thought of either a fork, or you know, it could be a subproject of Vitess... But then I felt that that will restrict its freedom right now. Starting it as an independent project -- these are very difficult decisions, and I think once we make thes... |
**Jerod Santo:** Yeah, I think you want the space to reinvent and rethink without feeling at all burdened by your past decisions with Vitess. |
**Sugu Sougoumarane:** Correct, correct. |
**Jerod Santo:** And I think -- obviously, copy and paste the stuff that makes sense... But don't make it a fork where you're like basically starting at this foundational, which you may end up wildly different in implementation. Or you might be the same, but at least give yourself room to come there organically, you kn... |
**Sugu Sougoumarane:** Yeah, that is something that was pretty obvious to me... Although the fundamentals are the same. Even if you don't copy code, you can copy ideas pretty easily. |
**Jerod Santo:** Sure. |
**Sugu Sougoumarane:** So those don't change. But when you get into the details, there's so many small things... And the thing is, if you don't do them the Postgres way, that will keep nagging you. You know, "I have written this... There's this if statement hanging there, which is there only to differentiate between Po... |
**Jerod Santo:** Right. |
**Sugu Sougoumarane:** You don't want any of that. |
**Jerod Santo:** None of that. |
**Sugu Sougoumarane:** Yeah. |
**Jerod Santo:** That kills the joy part, right? It takes away the joy that you're trying to enjoy what you guys are building here. Well, it's very exciting. Obviously, you're just getting started, so there's still a lot of question marks for us Postgres fans... It's just a lot of waiting, watching maybe the GitHub pro... |
**Sugu Sougoumarane:** The thing I feel is I feel a bit anxious that there's not much activity in the open source project. We may seem cool, but we are paddling hard underneath. Really, really hard. So I will be actually publishing a few blogs about some of the thoughts... There's one area which we didn't cover much of... |
\[01:05:58.20\] I feel like this is something -- the database world kind of went a little bit astray with mounted storage, disaggregated... I feel like that needs to be brought back in. I feel that storage should be with the database, because database are IOPS-hungry, and having your data local to your database is very... |
**Jerod Santo:** So are those things that Vitess tackled or did not tackle? High availability... |
**Sugu Sougoumarane:** Yeah, those are very core to Vitess. |
**Jerod Santo:** Yeah, I thought so. |
**Sugu Sougoumarane:** Those parts we are going to bring into Postgres. |
**Jerod Santo:** Right. Interesting. Adam, anything on your mind? Any other questions? |
**Adam Stacoviak:** I'm thinking about Neon, honestly. I'm thinking about just this kind of bigger picture... I'm curious if this shake-up, this change in the database world -- like, for a while there we had obviously PlanetScale leading, massively scale MySQL, which was great for MySQL, but not so great for the Postgr... |
**Sugu Sougoumarane:** Yeah, going back to why I started this project, right? The reason why I started this project was because I believed - I still believe, actually - that there is no scalability solution for Postgres. Not Neon, not Aurora. I'm an advisor to Metronome, which is a Postgres shop. They have one of the l... |
\[01:09:57.00\] And one of the things I thought with Supabase -- so this is a problem that is meant to solve it for the industry overall. In other words, Neon could use Multigres if they wanted. If they hit the limit of a machine, you could deploy Multigres on Neon and run it... Unless there were incompatibilities, for... |
And for Supabase specifically, is that if Supabase is going in its trajectory, there are going to be users that are going to max out a single instance very, very quickly. As a matter of fact, I won't be surprised if some are already hitting those limits. And their way out is you could go to Aurora maybe, and that will ... |
So the only way out of this is a sharded solution. And that is why I started this project. In MySQL, if you hit your limits, you could go to Vitess, but there is nothing like that in Postgres. And Vitess for Postgres, at even mediocre level of functionality, will allow these people to stay in Postgres. Otherwise, you h... |
**Jerod Santo:** Well, after you accomplish Vitess for Postgres, maybe you can build Vitess for SQLite. No, I'm just kidding. \[laughter\] |
**Adam Stacoviak:** He almost punched the screen, Jerod. He's like "I'm gonna punch my screen right now." |
**Jerod Santo:** After your next sabbatical, you know... The only thing that gets you back is that Dr. Richard Hipp approaches you and says "Sugu, we need you. We need you at SQLite." |
**Sugu Sougoumarane:** \[laughs\] That reminds me of -- I think Airplane, was it? ...where there's a background scene with a movie poster which says Rocky 83, and the really, really old guys \[unintelligible 01:12:24.23\] |
**Adam Stacoviak:** Airplane. You're going back in the day, into the '80s, to mention a movie... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.