[Top] [All Lists]

Re: [reiserfs-list] Re: benchmarks

To: Hans Reiser <reiser@xxxxxxxxxxx>
Subject: Re: [reiserfs-list] Re: benchmarks
From: Chris Wedgwood <cw@xxxxxxxx>
Date: Mon, 16 Jul 2001 21:00:29 +1200
Cc: rsharpe@xxxxxxxxxx, Xuan Baldauf <xuan--reiserfs@xxxxxxxxxxx>, Seth Mos <knuffie@xxxxxxxxx>, Federico Sevilla III <jijo@xxxxxxxxxxxxxxxxxxxx>, linux-xfs@xxxxxxxxxxx, reiserfs-list@xxxxxxxxxxx
In-reply-to: <3B5169E5.827BFED@xxxxxxxxxxx>
References: <Pine.BSI.4.10.10107141752080.18419-100000@xxxxxxxxxxxxx> <3B507259.4436853E@xxxxxxxxxxx> <20010715052116.E7056@xxxxxxxxxxxxx> <3B50D058.3090008@xxxxxxxxxx> <3B50CFF6.4BE90AB0@xxxxxxxxxxx> <20010715155758.G7624@xxxxxxxxxxxxx> <3B5169E5.827BFED@xxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mutt/1.3.18i
On Sun, Jul 15, 2001 at 02:01:09PM +0400, Hans Reiser wrote:

    Making the server stateless is wrong


    making the readdir a multioperation act is wrong

why? i have 3M directories... ar you saying clients should read the
whole things at once?

    but making not letting the FS use filename as a cookie and making
    it use 64 bit byte offsets is the most wrong thing of all.

here i agree, smarter clients could/should hang onto the filename and
revalidate if needed --- the only nasty things is REALLY deep
directory trees


<Prev in Thread] Current Thread [Next in Thread>