Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
My apologies for top posting, answering from blackberry. You're right about the actual numbers but I imagine when more spindles are involved, percentages will stay about the same. Best Regards, Artur eBoundHost http://www.eboundhost.com -----Original Message----- From: Trent Lloyd <lathiat at bur.st> Date: Thu, 25 Sep 2008 09:10:55 To: <eBoundHost at em.eboundhost.comArturartureboundhost.com> Cc: <drbd-user at lists.linbit.com> Subject: Re: [DRBD-user] Writes are much too slow Hi Artur, On 25/09/2008, at 8:05 AM, eBoundHost: Artur wrote: > I don't know if it's silly, seems to be the faster the better! Once > I have a better handle on this, we're going to be binding 3 or 4 > interfaces to remove a bottleneck. Who knows, a rack full of hard > drives may be just the thing for Dolphin. I think this is very cool > and necessary to take DRBD to the enterprise. > > To get back to what you're saying about the latency but take a look > at this: > http://lists.linbit.com/pipermail/drbd-user/2007-December/008128.html > > To my understanding we should be lagging by some percentage, but not > by such a margin. > In the email you posted they went from 13M/s to 4M/s so theyre seeing a large drop.. in fact by a larger factor than you were. Clearly your I/O subsystem is slower (in spins.. spindles.. not in raw throughput) than theirs because their raw speed is 13M vs 4M so thats where the difference in their numbers is I guess. I"m sure others may have some useful information on getting your latency down, thats not my area of expertise. Regards, -- Trent Lloyd -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20080925/0be86d42/attachment.htm>