News for hackers. Stuff that matters.

A comment on BlazingSQL – A GPU SQL Engine for RAPIDS Open-Source Software from Nvidia. Parent comment

Re:BlazingSQL – A GPU SQL Engine for RAPIDS Open-Source Soft... by usgroup on Friday October 12, 2018 @06:08AM (#18200395)
Personally I'm most excited about PGStrom, because it means faster Postgres all the time, not just when it fits into memory, and not just for SQL.
[ Parent | on HN ]
Re:BlazingSQL – A GPU SQL Engine for RAPIDS Open-Source Soft... by felipe_aramburu on Friday October 12, 2018 @10:29AM (#18201654)
Also to be clear as of right now the only thing that has to fit in memory are the inputs and the outputs to this system since these are being passed via CUDA IPC.
[ Parent | on HN ]
Re:BlazingSQL – A GPU SQL Engine for RAPIDS Open-Source Soft... by arnon on Friday October 12, 2018 @09:18AM (#18201189)
It appears to be, more than anything else, some CUDA code wrapped in UDFs. Not sure I'd want that anywhere near my production stuff in that format.
[ Parent | on HN ]
Re:BlazingSQL – A GPU SQL Engine for RAPIDS Open-Source Soft... by usgroup on Friday October 12, 2018 @09:43AM (#18201357)
This is false. PG has a well defined and modular query engine. pgstrom works by replacing and adding modules which allow for GPU support.
[ Parent | on HN ]
Re:BlazingSQL – A GPU SQL Engine for RAPIDS Open-Source Soft... by arnon on Friday October 12, 2018 @10:09AM (#18201505)
You're right, I got confused with another Postgres-GPU plugin type product.

However, Postgres isn't columnar. How does PG-Strom (now HeteroDB) arrange data for the GPU properly?

[ Parent | on HN ]