Re: Load distributed checkpoint - Mailing list pgsql-hackers

From ITAGAKI Takahiro
Subject Re: Load distributed checkpoint
Date
Msg-id 20061208134120.6658.ITAGAKI.TAKAHIRO@oss.ntt.co.jp
Whole thread Raw
In response to Re: Load distributed checkpoint  ("Takayuki Tsunakawa" <tsunakawa.takay@jp.fujitsu.com>)
Responses Re: Load distributed checkpoint
List pgsql-hackers
"Takayuki Tsunakawa" <tsunakawa.takay@jp.fujitsu.com> wrote:

> I'm afraid it is difficult for system designers to expect steady
> throughput/response time, as long as PostgreSQL depends on the
> flushing of file system cache. How does Oracle provide stable
> performance?
> Though I'm not sure, isn't it the key to use O_SYNC so that write()s
> transfer data to disk?

AFAIK, other databases use write() and fsync() in combination. They call
fsync() immediately after they write buffers in some small batches. Otherwise,
they uses asynchronous and direct I/O options. Therefore, dirty pages in
kernel buffers are keeped to be low at any time.

Regards,
---
ITAGAKI Takahiro
NTT Open Source Software Center




pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Load distributed checkpoint
Next
From: Gavin Sherry
Date:
Subject: Re: Weak passwords and brute force attacks