From owner-freebsd-questions Fri Mar 6 11:36:07 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA09192 for freebsd-questions-outgoing; Fri, 6 Mar 1998 11:36:07 -0800 (PST) (envelope-from owner-freebsd-questions@FreeBSD.ORG) Received: from stage1.thirdage.com (stage1.ThirdAge.com [204.74.82.151]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA08836 for ; Fri, 6 Mar 1998 11:35:01 -0800 (PST) (envelope-from jal@42is.com) Received: from goober (gigi.ThirdAge.com [204.74.82.169]) by stage1.thirdage.com (8.8.5/8.8.5) with SMTP id LAA02005; Fri, 6 Mar 1998 11:31:23 -0800 (PST) Message-Id: <3.0.3.32.19980306113103.00aa9180@colonel.42inc.com> X-Sender: jal@colonel.42inc.com X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Fri, 06 Mar 1998 11:31:03 -0800 To: Webmaster From: Jamie Lawrence Subject: Re: Webserver benchmarks? Ultra 2 vs. FreeBSD Cc: questions@FreeBSD.ORG In-Reply-To: <19980306164853.AAA27430@stimpy> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG At 08:49 AM 3/6/98 -0800, you wrote: >Hello all, > >I am running a Sun Ultra2 with Netscape Enterprise server 3.5.1, also have >P166 running 2.2.5-R and Apache 1.2.4. Both are on 100/Ethernet. > >Is there a good reason why the P166 would bench at about 60 requests/second >and the Ultra2 only about 15 requests/second? These tests were run using 3 >NT workstations as "web" clients. There are plenty of reasons why this could happen. You provide absolutely no information that could help one analyze why. How much RAM? What does iostat output look like? vmstat? How much time is your system spending in I/O wait? Have you looked at Netscape's performance tuning documents? Anything anomalous in netstat -s? In any case, this is a question for a list like sun-managers@ra.mcs.anl.gov, not Free-BSD Questions. We have several Ultra 2s in different configurations running Netscape Enterprise 2.0.1, and see better performance (by either one or two orders of magnitude, depending on what you're calling a "request"). >My employer insists that this is wrong. I used Web Bench 1.1 from ZDNet. People paying for expensive hardware usually don't like to see results like that, no. I can certainly believe those results, though, for a misconfigured box. -j To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message