LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

Re: Fundamental performance limits of a linux based load balancer /serve

To: "LinuxVirtualServer.org users mailing list." <lvs-users@xxxxxxxxxxxxxxxxxxxxxx>
Subject: Re: Fundamental performance limits of a linux based load balancer /server system?
From: Horms <horms@xxxxxxxxxxxx>
Date: Fri, 21 May 2004 12:19:44 +0900
On Tue, May 11, 2004 at 01:32:27PM -0700, Edward Chee wrote:
> I see. Pls excuse my ignorance - I'm new to all this :-)
> 
> Hmm. I wonder how this TCP port issue directly limits the Zeus software load
> balancer's incoming connection rate. In section 1.4 of
> http://support.zeus.com/doc/zlb/sizing_guide.pdf, they describe this
> problem.
> 
> Could it be because their balancer actually acts as a HTTP reverse proxy,
> accepting all connections and then having to actually act as a client to the
> real servers? I guess that must be it -- since their brochure at
> http://www.zeus.com/library/products/zlb/zlb_brochure.pdf indicates that
> they automatically translate HTTP 1.0 requests into HTTP 1.1 pipelined
> connections.

Yes, I would guess so to.

> To rephrase: A load balancer that operates on Layer 7 and maintains
> persistent connections to the real servers has to act as a TCP client and is
> subject to the port limitations. Is my thinking correct?

Yes. Though if there are multiple source addresses available, 
then it would have more ports available.

-- 
Horms
<Prev in Thread] Current Thread [Next in Thread>