Date: Wed, 11 Nov 1998 11:53:55 -0700 From: "Justin T. Gibbs" <gibbs@plutotech.com> To: Mike Smith <mike@smith.net.au> Cc: "Justin T. Gibbs" <gibbs@plutotech.com>, Greg Lehey <grog@lemis.com>, hackers@FreeBSD.ORG Subject: Re: SCSI vs. DMA33.. Message-ID: <199811111900.MAA04591@pluto.plutotech.com> In-Reply-To: Your message of "Wed, 11 Nov 1998 10:08:08 PST." <199811111808.KAA04141@dingo.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>Unfortunately, Simon's numbers tend to indicate that CAM doesn't >provide the same order of magnitude improvement that the old SCSI >subsystem did. At least it's a little more robust. 8) I never said it was order of magnitude improvement only an order of magnitude increase in concurrency. The numbers I've seen is on the order of 50% to 100% change in performance. Simon's benchmark is hitting the O(N) constraint of bufqdisksort which I'll fix as soon as I get a chance. I never received numbers from him with a change to remove the kernel elevator sort, but I would expect to see an improvement in scalability. -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199811111900.MAA04591>