postgres_vectorization_test
Data
database extension
SHOW HN2014
How HN took it
8 positive, 9 neutral or asking, 1 mixed, 3 critical, out of 21 comments about the project. Thread score +0.33.
Performance: 3 for, 0 againstIdea: 1 for, 2 againstExecution: 3 for, 0 againstDocs: 0 for, 1 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Impressive performance gains and well-written article
From 3 comments
@OP: A really good read and performance find. I wouldn't have expected those kinds of gains, but I wouldn't have considered the impact on CPU optimization. The improvements are…
Rapzid on HNWell done Can! keep up your work.
ceyhunkazel on HNAre you kidding? Articles like this are my favourite posts on HN. I would love to see more technical content like this and less softcore chit-chat.
DrJokepu on HN
Objections
Only applies to cstore_fdw, not core PostgreSQL
From 2 comments
If you read through it, this is a change to the Citus cstore_fdw extension, which stores data outside PostgreSQL and queries it using the PostgreSQL Foreign Data Wrapper API (see…
yangyang on HNI can't imagine this would ever get into core Postgres. It's a very specific extension.
yangyang on HN
Based on existing VectorWise ideas, not original and insufficiently credited
From 2 comments
Yes. Also, buried inside is the reference of the original work by VectorWise this work was based on. "Vectorization vs. Compilation in Query Execution". They don't even mention…
alecco on HNThey implemented ideas from a VectorWise paper. It's very good for an intern, but it's not original work.
alecco on HN
Requests
Bring the optimization into PostgreSQL core
From 2 comments
So what's the likelihood that we'll see this optimization make it into PostgreSQL core? That would be an amazing performance improvement.
lee on HNGreat article. I am not a real programmer (but regular user of Postgres for scientific data storage) and do not get every detail. But do these changes have a chance of being…
clemsen on HN
Questions
How does vectorized execution benefit from CPU features and cache?
From 1 comment
@OP: A really good read and performance find. I wouldn't have expected those kinds of gains, but I wouldn't have considered the impact on CPU optimization. The improvements are…
Rapzid on HN
Alternatives named
PGStrom (GPU) and PGOpenCL
From 1 comment
There are some other interesting PostgreSQL FDW extensions targeting speed. PGStrom uses the GPU: https://wiki.postgresql.org/wiki/PGStrom PGOpenCL:…
yangyang on HN
MonetDB and VectorWise
From 1 comment
Take a look at MonetDB, an open source column store database [1]. There was a commercialised extension to that called X100 targeting cache efficiency [2]. That became VectorWise…
yangyang on HN