Postgres Pro
Downloads
Home   >   mailing lists

RE: Protect syscache from bloating with negative cache entries - Mailing list pgsql-hackers

From Tsunakawa, Takayuki
Subject RE: Protect syscache from bloating with negative cache entries
Date
Msg-id 0A3221C70F24FB45833433255569204D1FB9DEA3@G01JPEXMBYT05
Whole thread Raw
In response to RE: Protect syscache from bloating with negative cache entries  ("Ideriha, Takeshi" <ideriha.takeshi@jp.fujitsu.com>)
Responses RE: Protect syscache from bloating with negative cache entries  ("Ideriha, Takeshi" <ideriha.takeshi@jp.fujitsu.com>)
List pgsql-hackers
Tree view 
From: Ideriha, Takeshi [mailto:ideriha.takeshi@jp.fujitsu.com]
> number of tables   | 100  |1000    |10000
> -----------------------------------------------------------
> TPS (master)       |10966  |10654 |9099
> TPS (patch)        | 11137 (+1%) |10710 (+0%) |772 (-91%)
> 
> It seems that before cache exceeding the limit (no pruning at 100 and 1000),
> the results are almost same with master but after exceeding the limit (at
> 10000)
> the decline happens.

How many concurrent clients?

Can you show the perf's call graph sampling profiles of both the unpatched and patched version, to confirm that the
bottleneckis around catcache eviction and refill?
 


Regards
Takayuki Tsunakawa

pgsql-hackers by date:

Previous
From: RSR999GMAILCOM
Date:
Subject: Using old master as new replica after clean switchover
Next
From: Michael Paquier
Date:
Subject: Re: Prepared transaction releasing locks before deregistering its GID