Re: [squid-users] Evaluating SQUID as a Reverse Proxy for a large web site

From: Koji Hino <hino@dont-contact.us>
Date: Sun, 07 Mar 2004 13:22:10 -0800 (PST)

Hi,

How you "benchmark" Apache and Squid?

I mean:

* Real environment benchmark, i.e. compare between followings with real
  customers
  - Get static contents from your busy (due to dynamic contents
    generation) Apache server
  - Get static contents from Squid, and get dynamic contents from your
    Apache server through Squid

OR

* Synthetic benchmark such as:
  - Get limited number (say several hundreds) of static contents, from
    your Apache server, or from your Squid server, with benchmarking
    HTTP client(s)

I think those two benchmarks are completely different things.

Best regards,

====================================================================
Koji HINO(HINO is my family name)
NEC Laboratories America, Inc.

From: "kapil khanna" <kapil@j2eerules.com>
 Date: Sun, 7 Mar 2004 13:13:43 -0600

:> Henrik,
:> I am using Apache as my web server. I decreased the dish cache size to 1 MB
:> and re-ran my tests. This time i got all TCP_MEM_HITS for the images and
:> static files. However my benchmark results were not better. The web server
:> still scaled much better.
:> My guess is that SQUID cannot handle concurrency too well. How do i get
:> SQUID to increase the no of processes or threads? Is my only option to have
:> many SQUID servers on different ports on a host with lot of RAM, running in
:> front of a load balancer to handle concurrency?
:> Thanks,
:> --Kapil
:> ----- Original Message -----
:> From: "Henrik Nordstrom" <hno@squid-cache.org>
:> To: "kapil khanna" <kapil@j2eerules.com>
:> Cc: <squid-users@squid-cache.org>
:> Sent: Sunday, March 07, 2004 3:28 AM
:> Subject: Re: [squid-users] Evaluating SQUID as a Reverse Proxy for a large
:> web site
:>
:>
:> > On Sat, 6 Mar 2004, kapil khanna wrote:
:> >
:> > > I have been evaluating SQUID to deploy in front of a large web site to
:> cache
:> > > all static content (Images, JS Files, CSS Files, HTML files etc...) for
:> the
:> > > web site. I used JMeter as a load testing tool to evaluate the
:> scalability
:> > > of SQUID. This is my current config:-
:> > > cache_mem - 256MB
:> > > disk cache - 10MB.
:> > > I purposely have a very low disk cache so that i can get most out of
:> > > in-memory caching of static content. I also set content expiry (if not
:> set)
:> > > for images, JSP files etc to
:> > > 14400 80% 43200
:> >
:> > You probably should run without any disk cache at all in this
:> > configuration. If not Squid will not actually be able to use all that
:> > cache_mem..
:> >
:> > > Why is that the Web application scales better than SQUID? The one thing
:> > > that stands out is that SQUID is running as one process one thread,
:> > > whereas the web site is multithreaded.
:> >
:> > What kind of web server are you using?
:> >
:> > The benefits of using a cache infront of the web server is mostly seen if
:> > the web server can not handle very many concurrent connections. The cache
:> > then helps both by offloading the static content any by reusing the same
:> > persistent connections for multiple clients.
:> >
:> > Regards
:> > Henrik
:> >
:> >
:>
Received on Sun Mar 07 2004 - 14:22:27 MST

This archive was generated by hypermail pre-2.1.9 : Thu Apr 01 2004 - 12:00:01 MST