From owner-freebsd-current Sat Apr 8 09:20:26 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA17962 for current-outgoing; Sat, 8 Apr 1995 09:20:26 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id JAA17956 for ; Sat, 8 Apr 1995 09:20:24 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.8/8.6.6) id JAA15432; Sat, 8 Apr 1995 09:19:53 -0700 From: "Rodney W. Grimes" Message-Id: <199504081619.JAA15432@gndrsh.aac.dev.com> Subject: Re: Disk performance To: taob@gate.sinica.edu.tw (Brian Tao) Date: Sat, 8 Apr 1995 09:19:53 -0700 (PDT) Cc: freebsd-current@FreeBSD.org In-Reply-To: from "Brian Tao" at Apr 8, 95 09:51:56 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 686 Sender: current-owner@FreeBSD.org Precedence: bulk > > On Sat, 8 Apr 1995, Gary Jennejohn wrote: > > > > 1st level cache: 407 MB/sec > [...] > > 1st level cache: 391 MB/sec > > Why would taking out the L2 cache slow down data transfer to and > from the primary cache? I attribute the above difference to an insignificant accuracy in the test results. (407-391)/391 = 4%, I suspect repeated runs of the same test on the same exact machine probably have a 5% variance in values. I see similiar variances with the program I use, from 227 to 263MB/sec or about 16% variation. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Custom computers for FreeBSD