Date: Thu, 13 Jul 2006 22:47:52 -0400 From: "Tamouh H." <hakmi@rogers.com> To: "'User Freebsd'" <freebsd@hub.org>, "'Jerry McAllister'" <jerrymc@clunix.cl.msu.edu> Cc: danial_thom@yahoo.com, 'FreeBSD Questions' <freebsd-questions@freebsd.org>, "'Chad Leigh -- Shire.Net LLC'" <chad@shire.net> Subject: RE: SMP Performance (Was: Re: Are hardware vendors starting to bail ... ) Message-ID: <20060714024743.BEC4B43D49@mx1.FreeBSD.org> In-Reply-To: <20060713220454.T1799@ganymede.hub.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> >> > >> On Jul 13, 2006, at 9:22 AM, Danial Thom wrote: > >> > >>> Simply enabling SMP on a single processor system adds 20-25%=20 > >>> overhead in freebsd 6.1. Again, readily admitted/accepted by the=20 > >>> developers. > >>> There is no way to recover that in efficiency, at least not for a=20 > >>> long time. > >> > >> So don't enable SMP on a single cpu system. Easy enough to avoid. > >> Chad > > > > Why would anyone want to enable SMP on a single CPU system anyway. >=20 > Actually, I believe all the new boot disks / ISOs are all=20 > SMP-enabled, so unless you build a custom kernel (some ppl do=20 > just run GENERIC ... I'm not one, mind you), you could be=20 > running an SMP-enabled kernel on a UP system without even=20 > knowing it ... >=20 > ---- > Marc G. Fournier Hub.Org Networking Services=20 > (http://www.hub.org) > Email . scrappy@hub.org MSN .=20 > scrappy@hub.org > Yahoo . yscrappy Skype: hub.org ICQ . 7615664 I have to put my two cents here: 1) I agree with few posters that FreeBSD performance have been lacking = behind. I've reported few issues on performance list and many did. We = offered few pre-production servers for performance testing, but the = answer we keep getting is: a. It is either your hardware sucks b. your benchmark application sucks 2) Regarding SMP, few posts talked about disabling hyper-thread and SMP = because it causes a performance degradation. On production hosting = server, the experience was otherwise though. Without HT and SMP, the = server would sky rocket in resource consumption. This has been tested on = FBSD 5.4 i386 3) I'm also frustrated like many with the rapid advancement in release = jumps. We barely started 5.x to conclude it does not live up to = expectations, so now 6.x is suppoused to be the good version, yet 7.x is = going to come out soon and probably in less than a year 6.x will be = considered inadequate. While I understand the development team is working hard to catch-up with = technology and hats off to all the developers, why there is "in my = opinion" no long term strategy for the base of the software. Thank you, Tamouh
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060714024743.BEC4B43D49>