RE: Global shared meta cache - Mailing list pgsql-hackers

From Tsunakawa, Takayuki
Subject RE: Global shared meta cache
Date
Msg-id 0A3221C70F24FB45833433255569204D1FA25E63@G01JPEXMBYT05
Whole thread Raw
In response to Global shared meta cache  ("Ideriha, Takeshi" <ideriha.takeshi@jp.fujitsu.com>)
List pgsql-hackers
From: Ideriha, Takeshi [mailto:ideriha.takeshi@jp.fujitsu.com]
> 1) Initialized with pgbench -i -s10
...
>    pgbench -c48 -T60 -Msimple      | 4956          |4965
> |95%

The scaling factor should be much greater than the number of clients.  Otherwise, multiple clients would conflict on
thesame row of branches table, which might hide the additional overhead of the global metacache.
 

And I think -j should be 12 or so on your 16-core server, so that the pgbench client can maximize its concurrency.


Regards
Takayuki Tsunakawa






pgsql-hackers by date:

Previous
From: Charles Cui
Date:
Subject: [GSoC] thrift compact protocol byte interface get implemented
Next
From: "Andrey V. Lepikhov"
Date:
Subject: Re: [WIP] [B-Tree] Retail IndexTuple deletion