Date: Fri, 22 Aug 1997 04:57:39 -0700 (PDT) From: asami@cs.berkeley.edu (Satoshi Asami) To: Shimon@i-Connect.Net Cc: tom@sdf.com, hackers@FreeBSD.ORG, mrcpu@cdsnet.net, nate@mt.sri.com Subject: Re: Final request for help with release. (DPT boot floppy) Message-ID: <199708221157.EAA00987@silvia.HIP.Berkeley.EDU> In-Reply-To: <XFMail.970821231148.Shimon@i-Connect.Net> (message from Simon Shapiro on Thu, 21 Aug 1997 23:11:48 -0700 (PDT))
next in thread | previous in thread | raw e-mail | index | archive | help
* I take this as a compliment. Thank you. How high is the load in your test. * We reject here anything under 256 concurrent processes and release after * 1024 concurrent processes ran continually for 48 hours. Then it goes to * System Test which does who know what... * * Simon * * Almost forgot; Still have to see a Solaris box survive this test. Let * alone perform as well as FreeBSD in these casual tests. I have also seen mysterious lockups (for a few seconds) with Solaris x86 when I bombarded more than two twin-channel controllers (Adaptec 3940W/UW) with lots of requests. FreeBSD had no problems whatsoever. Also, the maximum number of transactions (8KB random reads) per second of Solaris x86 maxed out at around 2,500, while FreeBSD went as high as 4,500 on identical hardware. This is with fast-wide SCSI -- with ultra-wide SCSI, FreeBSD went to 4,800 and Solaris couldn't run in ultra-wide mode. Also, FreeBSD could have been much higher if we used all twin-channel controllers, but we only used single-channel controllers because of the "identical hardware" requirement. This is with FreeBSD 2.2-stable of about half a year ago and Solaris x86 2.5.1 (the latast at that time). Satoshi
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199708221157.EAA00987>