Re: atime

From: John Kodis (kodis@gsfc.nasa.gov)
Date: Tue May 02 2000 - 09:49:13 EST


On Tue, 2 May 2000, Bruce Janson wrote:

> Maintenance of the traditional unix atime field turns read-like
> operations into write-like operations. For sophisticated network
> file systems that maintain a coherent distributed view, this
> hurts performance.
>
> Which popular/essential/useful user-level Linux applications
> break when atime is disabled (say, pegged to zero)?

You could eliminate most of the performance penalty by stealing an
idea that was used successfully on VMS: rather than simply pegging
atime to zero, provide an atime granularity mount option. With an
atime granularity of, say, 3600 (one hour), an atime update would only
be performed if not doing so would result in an atime value that was
in error by more than an hour. This limits atime updates to at most
one per hour per file.

-- 
John Kodis <kodis@acm.org>

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:10 EST