Subject: Re: yamt-readahead branch
To: None <feyrer@cs.stevens.edu>
From: YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp>
List: tech-kern
Date: 11/15/2005 17:11:28 
> Can you explain how readahead influences the random seeks/sec of 
> ~54 -> 69? is it because the readin's not that random at all?

current contextless read-ahead algorithm sometimes issues
read-ahead i/o even for random access.

YAMAMOTO Takashi