Date: Tue, 8 May 2007 17:22:19 -0700 From: Barrett Lyon <blyon@blyon.com> To: Kip Macy <kip.macy@gmail.com> Cc: adam radford <aradford@gmail.com>, freebsd-current@freebsd.org Subject: Re: Functional RAID controller? Message-ID: <AA1B8B95-0E73-4FDC-BB35-52EB1123AF3E@blyon.com> In-Reply-To: <b1fa29170705081612i43aedf70x8ba3e4a66f66380f@mail.gmail.com> References: <9FC464A4-4405-4C10-A7CB-0A424EA4EAD3@blyon.com> <b1bc6a000705081405s51c76ac4yf559bfd83affee8c@mail.gmail.com> <602A8820-F05C-457A-A20A-E258BD0FEDC5@blyon.com> <464102D1.2000706@samsco.org> <b1fa29170705081612i43aedf70x8ba3e4a66f66380f@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> In fairness, if you care about network bandwidth more than stability, > HEAD is the place to be. On my hardware if_mxge can get 9.3Gbps and > if_cxgb can get full line rate. if_mxge isn't even in RELENG_6 and > if_cxgb performance is at least 25% worse on RELENG_6. I can concur, that's why there is so much pressure to use HEAD, it's a substantial difference and all the network performance is found in HEAD, but it's useless if my disk arrays crash after writing some logs. :) -Barrett
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AA1B8B95-0E73-4FDC-BB35-52EB1123AF3E>