Date: Thu, 15 Jul 1999 11:37:23 -0700 From: "David Schwartz" <davids@webmaster.com> To: "Jamie Bowden" <ragnar@sysabend.org> Cc: "Doug" <Doug@gorean.org>, "Terry Lambert" <tlambert@primenet.com>, <scrappy@hub.org>, <beyssac@enst.fr>, <chat@FreeBSD.ORG> Subject: RE: Known MMAP() race conditions ... ? Message-ID: <000a01becef1$14003270$021d85d1@youwant.to> In-Reply-To: <Pine.BSF.3.96.990715083022.8686B-100000@beelzebubba.sysabend.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> On Wed, 14 Jul 1999, David Schwartz wrote: > > : > :> I can give you a list of things from my experience (not a > :> webmaster.com employee). Threads, SMP, NFS, and purify. I > can't think of > :> anything NT does better than any unix though. :) > : > : Large RAID arrays. 4-way SMP. Applications requiring large > numbers of > :threads. Log-based system. There's nothing I know of in any UNIX > that comes > :close to NT's completion ports for efficient network I/O. > > Irix can do all that, and use different familys and speeds in the same > box. Want to mix 200mhz R10k's and 250mhz R12k's in your Origin2k? Go > for it. We have a half a terrabyte RAID onsite at NASA LaRC, hanging off > an Irix box. That machine has more processors than NT can handle. Oh, > and it doesn't go down. Again, the question was "what does NT do better than _any_ UNIX", not "what does NT do better than _every_ UNIX". The only thing I know of that NT does better than _every_ UNIX is its I/O completion port mechanism. And those of you who are going to complain about things like where its buffers get handed off between user and kernel space simply don't understand that I/O completion ports are not about I/O, they're about thread management. DS To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?000a01becef1$14003270$021d85d1>