Thread: anyone tried to use hoard allocator?

Hi all,

today I've noticed this link on HN: http://plasma.cs.umass.edu/emery/hoard

Seems like an interesting option for systems with a lot of CPUs that are
doing a lot of alloc operations. Right now I don't have a suitable system
to test it - anyone tried to benchmark it?


Tomas

2012/3/26 Tomas Vondra <tv@fuzzy.cz>:
> Hi all,
>
> today I've noticed this link on HN: http://plasma.cs.umass.edu/emery/hoard
>
> Seems like an interesting option for systems with a lot of CPUs that are
> doing a lot of alloc operations. Right now I don't have a suitable system
> to test it - anyone tried to benchmark it?

It has sense for pg? It is not a multithread application.

Pavel

>
>
> Tomas
>
>
> --
> Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-performance

On Mar 26, 2012, at 2:50 AM, Tomas Vondra wrote:

> Hi all,
>
> today I've noticed this link on HN: http://plasma.cs.umass.edu/emery/hoard
>
> Seems like an interesting option for systems with a lot of CPUs that are
> doing a lot of alloc operations. Right now I don't have a suitable system
> to test it - anyone tried to benchmark it?

It's just another allocator - not a bad one, but it's been around for years. It's
mostly aimed at reducing contention in multi-threaded applications, so
it's not terribly applicable to strictly single-threaded postgresql.

It's licensing is pretty much incompatible with postgresql too.

Cheers,
  Steve