Re: Memory usage per session

On Fri, Jul 8, 2016 at 8:16 AM, <AMatveev(at)bitec(dot)ru> wrote:

> Hi
>
> > On 08/07/2016 14:11, AMatveev(at)bitec(dot)ru wrote:
> >> Hi.
> >> The test performs about 11K lines of code
> >> Memory usage per session:
> >> Oracle: about 5M
> >> MSSqlServer: about 4M
> >> postgreSql: about 160М
>
>
> > Visual C???
> > You will have to run PostgreSQL on a proper Unix system to test for
> performance.
> Of cause we understand that unix is faster( damn fork :)))
> Our Current problem is memory:
> (160m vs 5M) * 100 sessions = 16G vs 0.5G
> We just can get "out of memory".
> :(((
>

​I'm admittedly ignorant of this type of testing. But if the memory usage
for PostgreSQL is in the server, perhaps due to caching (how to test?),
then it likely would _not_ linearly scale up as the number of clients
increased because every clients would share the same cache data within the
server.​ Or are you measuring the memory usage where the client is on one
machine and the PostgreSQL server is a different machine, with the client
machine getting the memory hit?

--
"Pessimism is a admirable quality in an engineer. Pessimistic people check
their work three times, because they're sure that something won't be right.
Optimistic people check once, trust in Solis-de to keep the ship safe, then
blow everyone up."
"I think you're mistaking the word optimistic for inept."
"They've got a similar ring to my ear."

From "Star Nomad" by Lindsay Buroker:

Maranatha! <><
John McKown