back
user profile
levkk
2,540karma·349submissions·February 9, 2022
about
Sharding Postgres. <https://pgdog.dev>
lev@pgdog.dev
recent activity (349 total)
comment
Yeah, but I remember the plugin ignoring that setting.
comment
Last time I checked there was no support for synchronous logical replication in the `pglogical` plugin that's used for logical decoding. Did something change lately?
comment
I like. I had to do something similar a while back as part of an ETL pipeline.
comment
We're using persistent connections for Python/Gunicorn as well. See the Methodology section for more details.
comment
If I turn that into a single line and that improves performance 40x... I will probably not do engineering for a while after that.
comment
- We compared MessagePack as well, that's your typical binary format. It ended up being slower, which is what I've seen before when storing small floats (a typical ML feature). It's in …
comment
PostgresML v1.0 was doing exactly that. When we rewrote in Rust for v2.0, we improved 35x: https://postgresml.org/blog/postgresml-is-moving-to-rust-for... …
comment
We have LightGBM also. The entire Scikit library is available (although via a Python wrapper) and a couple algorithms from Linfa (a Rust Scikit rewrite). We will be adding more algorithms and move to …
comment
Link: https://postgresml.org/blog/postgresml-is-8x-faster-than-pyt... …
comment
My bad. Here it is: https://postgresml.org/blog/postgresml-is-8x-faster-than-pyt... …
comment
┌──(rootsf-BrushFocus)-[~]
└─# finger
No one logged on.
Where is everyone?
comment
Rewrite of PostgresML in Rust, with optimized XGBoost, LightGBM, Linfa, and vector operations. Fully backwards compatible to v1.0 and continues to support Python ML libraries, incl. scikit-learn.
comment
I had a great experience with Hotwired Turbo and Stimulus.
comment
Correct! According to Google, "insecure" means: 1. not firmly fixed; liable to give way or break.
2. (of a person) not confident or assured; uncertain and anxious. and "unsecure": …
comment
I used pg_repack at a very large scale, and one issue I ran into was write amplification. Since it used triggers to keep a log of rows changed, writes during repack were 2x normal. pg_squeeze use of r…