Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Better JIT for Postgres (github.com/vladich)
120 points by vladich 10 hours ago | hide | past | favorite | 49 comments
 help



We have everything optimized, and yet somehow DB queries need to be "interpreted" at runtime. There's no reason for DB queries to not be precompiled.

The "byte-code" coming from the query planner typically only has a handful of steps in a linear sequence. Joins, filters, and such. But the individual steps can be very costly.

So there is not much to gain from JITing the query plan execution only.

JITing begins to make more sense, when the individual query plan steps (join, filter, ...) themselves be specialized/recompiled/improved/merged by knowing the context of the query plan.


This is a neat idea. I want to take it further and precompile the entire DBMS binary for a specific schema.

How will you handle ALTER TABLE queries without downtime?

That would definitely present a bit of a challenge, but:

- not all databases need migrations (or migrations without downtime)

- alternatively, ship the migrations as part of the binary

Adhoc modifications would still be more difficult but tbh that’s not necessarily a bug


Many SQL engines have JIT compilers.

The problems related to PostgreSQL are pretty much all described here. It's very difficult to do low-latency queries if you cannot cache the compiled code and do it over and over again. And once your JIT is slow you need a logic to decide whether to interpret or compile.

I think it would be the best to start interpreting the query and start compilation in another thread, and once the compilation is finished and interpreter still running, stop the interpreter and run the JIT compiled code. This would give you the best latency, because there would be no waiting for JIT compiler.


> It's very difficult to do low-latency queries if you cannot cache the compiled code

This is not too difficult, it just requires a different execution style. Salesforce's Hyper for example very heavily relies on JIT compilation, as does Umbra [1], which some people regard as one of the fastest databases right now. Umbra doesn't cache any IR or compiled code and still has an extremely low start-up latency; an interpreter exists but is practically never used.

Postgres is very robust and very powerful, but simply not designed for fast execution of queries.

Disclosure: I work in the group that develops Umbra.

[1]: https://umbra-db.com/


> I think it would be the best to start interpreting the query and start compilation in another thread

This technique is known as a "tiered JIT". It's how production virtual machines operate for high-level languages like JavaScript.

There can be many tiers, like an interpreter, baseline compiler, optimizing compiler, etc. The runtime switches into the faster tier once it becomes ready.

More info for the interested:

https://ieeexplore.ieee.org/document/10444855


See prepared statements.

Postgresql uses a process per connection model and it has no way to serialize a query plan to some form that can be shared between processes, so the time it takes to make the plan including JIT is very important.

Most other DB's cache query plans including jitted code so they are basically precompiled from one request to the next with the same statement.


What do you mean ? Cause the obvious thing is a shared cache and if there is one thing the writers of a db know it is locking

Sharing executable code between processes it not as easy as sharing data. AFAIK unless somethings changed recently PG shares nothing about plans between process and can't even share a cached plan between session/connections.

Write the binary to a file, call it `libquery-id1234.so`, and link that to whichever processes that need it?

Might want to take a look at some research like this [1] that goes over the issues:

"This obvious drawback of the current software architecture motivates our work: sharing JIT code caches across applications. During the exploration of this idea, we have encountered several challenges. First of all, most JIT compilers leverage both runtime context and profile information to generate optimized code. The compiled code may be embedded with runtime-specific pointers, simplified through unique class-hierarchy analysis, or inlined recursively. Each of these "improve- ments" can decrease the shareability of JIT compiled code."

Anythings doable here with enough dev time. Would be nice if PG could just serialize the query plan itself maybe just as an SO along with non-process specific executable code that then has to be dynamically linked again in other processes.

1. https://dl.acm.org/doi/10.1145/3276494


Executable code is literally just data that you mark as executable. It did the JIT code, and the idea that it can't then share it between processes is incomprehensible.

I was actually confused by this submission as it puts so much of an emphasis on initial compilation time, when every DB (apparently except for pgsql) caches that result and shares it/reuses it until invalidation. Invalidation can occur for a wide variety of reasons (data composition changing, age, etc), but still the idea of redoing it on every query, where most DBs see the same queries endlessly, is insane.


The emphasis on compilation time there is because the JIT provider that comes with Postgres (LLVM-based) is broken in that particular area. But you're right, JITed code can be cached, if some conditions are met (it's position independent, for one). Not all JIT providers do that, but many do. Caching is on the table, but if your JIT-compilation takes microseconds, caching could be rather a burden in many cases. Still for some cases useful.

No a lot of jitted code has pointers to addresses specific to that process which makes no sense in another process.

To make code shareable between processes takes effort and will have tradeoff in performance since it is not specialized to the process.

If the query plan where at least serializable which is more like a AST then at least that part could be reused and then maybe have jitted code in each processes cached in memory that the plan can reference by some key.

DB's like MSSQL avoid the problem because they run a single OS process with multiple threads instead. This is also why it can handle more connections easily since each connection is not a whole process.


> and it has no way to serialize a query plan to some form that can be shared between processes

https://www.postgresql.org/docs/current/parallel-query.html

"PostgreSQL can devise query plans that can leverage multiple CPUs in order to answer queries faster."


Nothing to do with plan caching, thats just talking about plan execution of parallel operations which is that thread or process based in PG?

If process based then they can send small parts of plan across processes.


Ah, didn't see the caching part.

Plans for prepared statements are cached though.


Yes if the client manually prepares the statement it will be cached for just that connection because in PG a connection is a process, but it won't survive from one connection to the next even in same process.

Other databases like MSSQL have prepared statements but they are rarely used now days since plan caching based on query text was introduced decades ago.


Only on a per-connection basis

> However, standard LLVM-based JIT is notoriously slow at compilation. When it takes tens to hundreds of milliseconds, it may be suitable only for very heavy, OLAP-style queries, in some cases.

I don't know anything here, but this seems like a good case for ahead of time compilation? Or at least caching your JIT results? I can image much of the time, you are getting more or less the same query again and again?


Yes.

Some years ago we ported some code from querying out the data and tallying in Python (how many are in each bucket) to using SQL to do that. It didn't speed up the execution. I was surprised by that, but I guess the Postgres interpreter is roughly the same speed as Python, which when you think about it perhaps isn't that surprising.

But Python is truly general purpose while the core query stuff in SQL is really specialized (we were not using stored procedures). So if Pypy can get 5x speedup, it seems to me that it should be possible to get the same kind of speed up in Postgres. I guess it needs funding and someone as smart as the Pypy people.


At some level the application needs to participate in the performance conversation too.

https://www.postgresql.org/docs/current/sql-prepare.html


Postgres’s PREPARE is per-connection so it’s pretty limited, and then connection poolers enter the fray and often can’t track SQL-level prepares.

And then the issue is not dissimilar to Postgres’s planner issues.


Unless you cache query plans like other RDBMS's then the client manually managing that goes away and its not limited to a single connection.

MS SQL still has prepared statements and they really haven't been used in 20 years since it gained the ability to cache plans based on statement text.


The last time I looked into it my impression was that disabling the JIT in PostgreSQL was the better default choice. I had a massive slowdown in some queries, and that doesn't seem to be an entirely unusual experience. It does not seem worth it to me to add such a large variability to query performance by default. The JIT seemed like something that could be useful if you benchmark the effect on your actual queries, but not as a default for everyone.

That is quite strange, given that big boys RDMS (Oracle, SQL Server, DB2, Informix,...) all have JIT capabilities for several decades now.

The big boys all cache query plans so the amount it time it take to compile is not really a concern.

Postgres caches query plans too, the problem is you can only cache what you can share, and if your planner works well, you can share very little, there can be a lot of unique plans even for the same query

That's not generally correct. Compile-time is a concern for several databases.

Most systems submit many of the same queries over and over again.

Ad-hoc one off queries usually can accept higher initial up-front compile cost because the main results usually take much longer anyway, vs worrying about an extra 100ms of compile.

Maybe it was too strong to say its not a concern at all, but nothing like PG where every single request needs to replan and potentially jit unless the client manually prepares and keeps the connection open.


What sort of things are people doing in their SQL queries that make them CPU bound? Admittedly I'm a meat-and-potatoes guy, but I like mine I/O bound.

Really amazed to see not one but several generic JIT frameworks though, no idea that was a thing.


Most databases in practice are sub-terabyte and even sub-100Gb, their active dataset is almost fully cached. For most databases I worked with, cache hit rate is above 95% and for almost all of them it's above 90%. In that situation, most queries are CPU-bound. It's completely different from typical OLAP in this sense.

Anything jsonb in my experience is quickly CPU bound...

Definitely. If you're doing regular queries with filters on jsonb columns, having the index directly on the JSON paths is really powerful. If I have a jsonb filter in the codebase at all, it probably needs an index, unless I know the result set is already very small.

I think reading queries that are always served from cache are CPU bound because it also involves locking the buffers etc and there is no I/O involved.

PostgreSQL is Turing complete, so I guess they do what ever they want?

Have you tested this under high concurrency with lots of short OLTP queries? I’m curious whether the much faster compile time actually moves the point where JIT starts paying off, or if it’s still mostly useful for heavier queries.

> By default, jit_above_cost parameter is set to a very high number (100'000). This makes sense for LLVM, but doesn't make sense for faster providers. It's recommended to set this parameter value to something from ~200 to low thousands for pg_jitter (depending on what specific backend you use and your specific workloads).

sadly, no windows version yet AFAICT

awesome! I wonder if it's possible to point AI at this problem and synthesize a bespoke compiler (per-architecture?) for postgresql expressions?

Two things are holding back current LLM-style AI of being of value here:

* Latency. LLM responses are measured in order of 1000s of milliseconds, where this project targets 10s of milliseconds, that's off by almost two orders of magnitute.

* Determinism. LLMs are inherently non-deterministic. Even with temperature=0, slight variations of the input lead to major changes in output. You really don't want your DB to be non-deterministic, ever.


> LLMs are inherently non-deterministic.

This isn't true, and certainly not inherently so.

Changes to input leading to changes in output does not violate determinism.


> This isn't true

From what I understand, in practice it often is true[1]:

Matrix multiplication should be “independent” along every element in the batch — neither the other elements in the batch nor how large the batch is should affect the computation results of a specific element in the batch. However, as we can observe empirically, this isn’t true.

In other words, the primary reason nearly all LLM inference endpoints are nondeterministic is that the load (and thus batch-size) nondeterministically varies! This nondeterminism is not unique to GPUs — LLM inference endpoints served from CPUs or TPUs will also have this source of nondeterminism.

[1]: https://thinkingmachines.ai/blog/defeating-nondeterminism-in...


Quoting:

"But why aren’t LLM inference engines deterministic? One common hypothesis is that some combination of floating-point non-associativity and concurrent execution leads to nondeterminism based on which concurrent core finishes first."

From https://thinkingmachines.ai/blog/defeating-nondeterminism-in...


> 1000s of milliseconds

Better known as "seconds"...


The suggestion was not to use an LLM to compile the expression, but to use an LLM to build the compiler.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: