From owner-freebsd-stable Thu Oct 25 14:29:50 2001 Delivered-To: freebsd-stable@freebsd.org Received: from salmon.maths.tcd.ie (salmon.maths.tcd.ie [134.226.81.11]) by hub.freebsd.org (Postfix) with SMTP id 30AB637B403 for ; Thu, 25 Oct 2001 14:29:47 -0700 (PDT) Received: from walton.maths.tcd.ie by salmon.maths.tcd.ie with SMTP id ; 25 Oct 2001 22:29:43 +0100 (BST) To: Sean Chittenden Cc: Ian Dowse , Cy Schubert - ITSD Open Systems Group , Anatoliy Dmytriyev , freebsd-stable@FreeBSD.ORG Subject: Re: UFS_DIRHASH - your opinion In-reply-to: Your message of "Thu, 25 Oct 2001 14:06:51 PDT." <20011025140651.A8755@ninja1.internal> X-Request-Do: Date: Thu, 25 Oct 2001 22:29:43 +0100 From: David Malone Message-ID: <200110252229.aa61044@salmon.maths.tcd.ie> Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG > How about for Maildir mailboxes? In anyone mailbox I have roughly > 10,000 messages which means 10,000 files per directory and mutt > opening/reading the directory every time I switch mail folders. I suspect that you'll see a big improvement with directories like this. > Granted I should rebuild a system and do the tests myself, but do you > have any preliminary numbers in terms of slowdowns for small dirs, and > speedups for large dirs? -sc There should be no loss of performance for small directories. The main area where there might be a loss would be if your application just looks up one entry in the directory and doesn't return to that directory again. BTW - you don't need to rebuild the filesystem to use dirhash. All you need to do is compile it into the kernel. If it is compiled in then it automatically works on all ufs filesystems. David. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message