Date: Fri, 07 May 1999 00:47:58 +1200 From: Andrew McNaughton <andrew@squiz.co.nz> To: Andy Angrick <angrick@netdirect.net> Cc: Graeme Tait <graeme@echidna.com>, freebsd-isp@FreeBSD.ORG Subject: Re: Apache Stress testing Message-ID: <199905061247.AAA04038@aniwa.sky> In-Reply-To: Your message of "Tue, 04 May 1999 10:09:58 EST." <1.5.4.32.19990504150958.00bfb424@netdirect.net>
next in thread | previous in thread | raw e-mail | index | archive | help
> Agreed. It would have to have some way of simulating longer connection > times. Like coming from a 28.8 modem or whatever. Maybe even a randomly > selected stay online time. Would rigging it up to go through dummy net pipes fit the bill perhaps? I'm sure a more complex simulation could be dreamed up, but as a simply implemented simulation of narrow bandwidth connections it has some merit. In my use, apache's repsonsiveness is significantly affected by system load, but (excepting external processes like cgi) doesn't actually create much of the load on my system. One query I have: if my file system keeps tabs on when files were last accessed, does that mean that every file opened requires a disk access, even if it is cached, or does this mark the cached info as dirty so that it will be written at a future time? Does anyone have any experience with the effect on apache performance of mounting the file system with the noatime flag? Andrew McNaughton -- ----------- Andrew McNaughton andrew@squiz.co.nz http://www.newsroom.co.nz/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199905061247.AAA04038>