Date: Fri, 13 Oct 2006 12:39:00 +0800 From: "Adrian Chadd" <adrian@freebsd.org> To: "Mark Linimon" <linimon@lonesome.com> Cc: FreeBSD Stable <freebsd-stable@freebsd.org>, Kris Kennaway <kris@obsecurity.org> Subject: Re: [fbsd] HEADS UP: FreeBSD 5.3, 5.4, 6.0 EoLs coming soon Message-ID: <d763ac660610122139i1c03695ex597802c08af84dd8@mail.gmail.com> In-Reply-To: <20061013030825.GB17131@soaustin.net> References: <451F6E8E.8020301@freebsd.org> <20061011102106.GY1594@obiwan.tataz.chchile.org> <20061011151458.L97038@atlantis.atlantis.dp.ua> <20061011083021.C2780@treehorn.dfmm.org> <452D6CFD.4040905@FreeBSD.org> <20061011223610.GA30707@riyal.ugcs.caltech.edu> <08C06C77-F9CB-4545-8A68-9E930E2C58DD@khera.org> <20061012184845.GA59291@xor.obsecurity.org> <20061013030825.GB17131@soaustin.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 10/13/06, Mark Linimon <linimon@lonesome.com> wrote: > > DragonFly has made substantial rewrites/changes since the fork from > FreeBSD. > I think to assume that there are no regressions in either stability, > speed, > or support may be naive. Has anyone tried benchmarking DragonflyBSD against FreeBSD 5.x and 6.x for some of the specific workloads people are reporting issues with? I'll try to get around to benchmarking TCP socket transaction throughput (connect()/accept(), read/write, close() as FreeBSD 6 and Linux-2.6 systems seem to spend quite a large percentage of CPU time creating and tearing down TCP sockets for some reason. No, I haven't yet done any kernel profiling, this is just from measured transaction rates from userland. Adrian -- Adrian Chadd - adrian@freebsd.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d763ac660610122139i1c03695ex597802c08af84dd8>