From owner-freebsd-chat Tue Oct 24 20:58: 0 2000 Delivered-To: freebsd-chat@freebsd.org Received: from prism.flugsvamp.com (cb58709-a.mdsn1.wi.home.com [24.17.241.9]) by hub.freebsd.org (Postfix) with ESMTP id 53A2037B4C5 for ; Tue, 24 Oct 2000 20:57:58 -0700 (PDT) Received: (from jlemon@localhost) by prism.flugsvamp.com (8.11.0/8.11.0) id e9P3ubh54849; Tue, 24 Oct 2000 22:56:37 -0500 (CDT) (envelope-from jlemon) Date: Tue, 24 Oct 2000 22:56:37 -0500 From: Jonathan Lemon To: Dan Kegel Cc: chat@freebsd.org Subject: kqueue Message-ID: <20001024225637.A54554@prism.flugsvamp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre2i Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I recently stumbled across a message you posted asking for microbenchmarks on kqueue. While I do think that microbenchmarks are partially misleading, I did run them on my machine for various numbers of connections, with varying number of active connections. The results are shown below. The results dovetail with what I expect: kqueue scales depending on the number of active connections that it sees, not with the total number of connections. Also, I presented a paper/talk at the recent BSDCon 2000, these are available at http://www.freebsd.org/~jlemon if you're interested. -- Jonathan This is on a single processor 600Mhz Pentium-III with 512MB of memory, running FreeBSD 4.x-STABLE: cache[10:13pm]> ./Poller_bench 5 1 spk XXX pipes 100 1000 10000 30000 select 54 - - - poll 50 552 11559 35178 kqueue 8 8 8 8 cache[10:13pm]> ./Poller_bench 5 10 spk XXX pipes 100 1000 10000 30000 select 100 - - - poll 95 571 11697 35499 kqueue 52 52 55 56 cache[10:13pm]> ./Poller_bench 5 50 spk XXX pipes 100 1000 10000 30000 select 299 - - - poll 287 887 12124 36054 kqueue 285 295 313 330 cache[10:13pm]> ./Poller_bench 5 100 spk XXX 14 microseconds to open each of 100 socketpairs pipes 100 1000 10000 30000 select 542 - - - poll 528 1091 12440 36530 kqueue 574 592 623 702 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message