Re: Is async the answer - Akins, Brian - org.apache.httpd.dev

21 messages in org.apache.httpd.devRe: Is async the answer

From	Sent On	Attachments
Akins, Brian	18 Jan 2008 10:52
Justin Erenkrantz	18 Jan 2008 11:16
Colm MacCarthaigh	18 Jan 2008 11:20
Akins, Brian	18 Jan 2008 11:27
Akins, Brian	18 Jan 2008 11:31
Colm MacCarthaigh	18 Jan 2008 12:07
Akins, Brian	18 Jan 2008 13:17
Colm MacCarthaigh	18 Jan 2008 13:29
Ruediger Pluem	18 Jan 2008 14:30
Justin Erenkrantz	18 Jan 2008 16:33
Niklas Edmundsson	19 Jan 2008 03:53
Graham Leggett	19 Jan 2008 04:45
Davi Arnaut	19 Jan 2008 06:57
Jim Jagielski	19 Jan 2008 07:04
Graham Leggett	19 Jan 2008 08:02
Henrik Nordström	19 Jan 2008 13:14
Henrik Nordström	19 Jan 2008 13:19
Davi Arnaut	19 Jan 2008 15:29
Graham Leggett	20 Jan 2008 07:44
Akins, Brian	22 Jan 2008 10:03
Akins, Brian	22 Jan 2008 10:08

Subject:	Re: Is async the answer
From:	Akins, Brian (Bria...@turner.com)
Date:	01/18/2008 10:52:02 AM
List:	org.apache.httpd.dev

On 1/18/08 12:18 PM, "Colm MacCarthaigh" <co...@allcosts.net> wrote:

Hmmm, it depends what you mean by scale really. Async doesn't help a daemon scale in terms of concurrency or throughput, if anything it might even impede it, but it certainly can help improve latency and responsivity greatly. On the whole, it's easy to see how it might make the end user experience of a very busy server much more pleasant.

I also wonder is that has actually been tested or if it's just a "factoid"?

Response time never increased in any measurable amount.

I suspect it might though if the scheduler became bound, async would route the interupts more efficiently.

But, I wonder if the scheduler would become bound in a "reasonable" amount of traffic.

discussions on scalability baffling, the reality is that modern hardware can outscale pretty much any amount of bandwidth you can buy regardless of the software.

Bandwidth generally isn't an issue for us anymore (thanks to gzip). We can still overrun the CPU with small objects requests/responses. On "large" objects (ie, over 16k or so), the CPU is bored when multiple gig interfaces are full.

The scalability wars should really be over, everyone won - kernel's rule :-)

Which is why I hate to see a ton of work go into async core if it actually does very little to help performance (or if it hurts it) and makes writing modules harder. It braindead simple nowadays to write well behaved high performance modules (well, mostly) bcs you rarely worry about threads, reads/writes, etc. Full async programming is just as challenging as handling a ton of threads yourself.

My $.02 US worth (which ain't much).