atom feed25 messages in ru.sysoev.nginxRe: A hardware question
FromSent OnAttachments
mikeApr 27, 2008 12:36 am 
Dave CheneyApr 27, 2008 1:21 am 
mikeApr 27, 2008 1:58 am 
Dave CheneyApr 27, 2008 2:11 am 
mikeApr 27, 2008 1:34 pm 
Janko HauserApr 27, 2008 1:44 pm 
Igor SysoevApr 27, 2008 9:34 pm 
mikeApr 27, 2008 10:52 pm 
Igor SysoevApr 27, 2008 11:05 pm 
Chavelle VincentApr 28, 2008 3:16 am 
Igor SysoevApr 28, 2008 7:10 am 
mikeApr 28, 2008 11:45 am 
kinglerApr 28, 2008 12:03 pm 
mikeApr 28, 2008 12:26 pm 
Ezra ZygmuntowiczApr 29, 2008 3:11 pm 
mikeApr 29, 2008 4:55 pm 
mikeApr 29, 2008 7:25 pm 
mikeApr 29, 2008 9:11 pm 
Manlio PerilloApr 30, 2008 1:11 am 
mikeApr 30, 2008 2:09 am 
Manlio PerilloMay 1, 2008 2:47 am 
mikeMay 1, 2008 8:17 am 
mikeMay 1, 2008 3:11 pm 
Eden LiMay 1, 2008 6:31 pm 
mikeMay 1, 2008 7:03 pm 
Subject:Re: A hardware question
From:mike (mike@public.gmane.org)
Date:Apr 27, 2008 1:58:11 am
List:ru.sysoev.nginx

physical cache being L2/L3 cache type stuff?

yeah i shouldn't be hitting the SATA bottleneck. right now most is served via NFS, hopefully migrating it to local disk for scripts/templates, and for data using mogilefs (which will run on these same machines)

On 4/27/08, Dave Cheney <dave-7L4Cwp9BzA+sTnJN9+BG@public.gmane.org> wrote:

The single SATA disk will be your bottleneck, followed by system time. I'd reserve the machines with the largest physical cache for php, its where you will see the biggest payoff.

IMHO

On 27/04/2008, at 5:36 PM, mike wrote:

I plan on growing up to 2 million php requests per day (per machine) and maybe 5 or 6 million static files (video, image, HTML, etc)