News for hackers. Stuff that matters.

jonahharris

Show HN: uplpgsql – PL/pgSQL compiled to native code

https://github.com/nextgres/uplpgsql

uplpgsql compiles PL/pgSQL to native code. It took over twenty years to get here, but here it is.On July 19, 2005, I demoed my optimizing PL/SQL compiler to Andy Astor, Denis Lussier, and Jim Mlodgenski at EnterpriseDB; PL/SQL packages, procedures, and functions ran on Postgres at native speed, not interpreted. While they didn't acquire it, they brought me on as a founding engineer focused on...

( Read More... | no comments | 3 points | on HN )
Many experienced founders get rejected by YC and blame ageism. I used to think that too.

After reflecting on my own experience, and talking with others, I think YC is mostly right.

Early-stage startups reward speed, not caution. Experience often slows founders down at the exact moment when reckless momentum matters most.

Here’s my breakdown of where/why older founders struggle and how to approach it more clearly.

Right now, the source is available under the Confluent license, but we intend to go BSD/MIT once our fundraising is done.

Show HN: New Postgres Connection Pooling Extension Breaks 100k Barrier

https://medium.com/@jonah.harris/surpassing-100k-postgres-connections-with-our-in-database-connection-pooling-extension-7084552234ea

If you’ve ever dealt with scaling PostgreSQL, you know the traditional process-per-connection model just doesn’t cut it when you're handling massive numbers of connections, particularly short-lived ones. When I led architecture at The Meet Group, which had one of the largest deployments of Postgres in the world, we optimized everything: connection pools in the app tier, as a man-in-the-middle,...

( Read More... | 3 comments | 5 points | on HN )
Re:The Future of MySQL is Postgres [pdf] on Friday April 26, 2024 @04:46PM (#40174080)
Based on conversations with a lot of developers and businesses, a few things happened, I think. First, MySQL hasn’t been keeping pace with Postgres feature wise; second, MySQL isn’t the migrated-to database from Oracle Database or SQL Server, Postgres is - and as companies don’t want to have to manage multiple databases, if their departmental apps all go to Postgres, then it becomes the default development database also. The most important, I think, is the MySQL community and ecosystem. There’s a ton of users and tools, but not much in terms of development or plugins that enhance the server like Postgres has. MySQL is still used heavily for high performance web sites but, aside from that, it’s not great in mixed workload OLTP or OLAP, and large OLAP workloads really need something like HeatWave. TiDB and Vitess are great for scale out OLTP when normal single server MySQL gets overwhelmed.
Re:The Future of MySQL is Postgres [pdf] on Saturday April 20, 2024 @12:58AM (#40094941)
From an embedded perspective or where the iOS/Android app is a client?
Re:The Future of MySQL is Postgres [pdf] on Friday April 19, 2024 @10:30PM (#40094160)
Postgres is on the rise. MySQL is in decline. But there are a helluva lot of MySQL applications out there and developers still building against it. While the industry is moving away from MySQL to Postgres, the only solution for this is traditional migrations, which--let's face it--are manageable from a data standpoint, but a pain in the ass to address with applications.

Today, at Postgres Conference 2024, we introduced a PostgreSQL extension we are developing, which enables MySQL applications to run on PostgreSQL without any code changes by supporting the MySQL wire protocol, SQL syntax, and procedural language. It addresses MySQL's large database performance issues by leveraging PostgreSQL's features, such as enhanced partitioning, analytical queries, parallel query, advanced indexing, and materialized views. The extension also gives MySQL developers access to the comprehensive PostgreSQL extension ecosystem, including columnar storage, vector database functionality, and in-database machine learning, providing additional functionality. I mean, Oracle just released a preview of JavaScript procedural language support in MySQL in 2024, whereas our extension inherits Postgres' PLs, giving MySQL developers access to in-database Python, Rust, JavaScript, et al.

Not only that, but our solution consolidates database management, eliminating the need to run both MySQL and PostgreSQL systems, which is common among 75% of businesses using PostgreSQL. This approach improves productivity and simplifies operations.

Our Alpha will be released in July, Beta in September, and we expect to be GA by EOY. Questions welcomed.

Re:Mongo Vs Redis, The Increment Battle on Friday January 14, 2011 @01:09PM (#2104645)
In his test, it appears the Mongo driver is just putting the message on the wire, not actually waiting on the server reply or verifying that the command was ever actually processed. Using the Redis in pipelined execution mode would be a much closer test.