Re: [squid-users] url_rewrite_concurrency singlethreaded redirector performance?

From: Sylvain Viart <sylvain.viart@dont-contact.us>
Date: Wed, 10 Oct 2007 10:19:54 +0200

Hi Henrik,

Thanks for the detailed reply.

Henrik Nordstrom a écrit :
> On tis, 2007-10-09 at 11:21 +0200, Sylvain Viart wrote:
>
>
>> Would than mean we have some performance issue with singlethreaded
>> redirector?
>>
>
> In the concurrent mode helper interface Squid batches up to the
> concurrency level number of requests to the same helper, making the
> communication squid<->helper much more efficient. This is actually
> unrelated of if the helper as such is multithreaded or singlethreaded.
>
I see, is it documented somewhere?
If I understand it well, given the following perl helper code:

while(<STDIN>)
{
    s#http://something/#http://somthingelse/#;
    print;
 }

can handle many linesof url request in the same loop, without any
multithreaded mechanism.

> Most existing helpers can easily be converted to the concurrent mode
> interface to benefit from the batching. The requirements is
> * That the helper uses proper line-oriented reading of requests, making
> sure that the next request is not forgotten if then are received at the
> same time.
>
I'm not sure it answer my previous question, but it seems it is.
> * The format of request/responses is slightly different from the older
> interface, adding a request identifier in front. This needs to be echoed
> back with the answer.
>
Documentation please?
Google doesn't seems to know anything interesting on url_rewrite_concurrency

Sylvain.
Received on Wed Oct 10 2007 - 02:20:02 MDT

This archive was generated by hypermail pre-2.1.9 : Thu Nov 01 2007 - 13:00:01 MDT