On Thu, Feb 22, 2001 at 03:01:55PM -0500, tc lewis wrote:
> On Thu, 22 Feb 2001, Alex Samonte wrote:
> > On Thu, Feb 22, 2001 at 03:44:44AM -0500, tc lewis wrote:
> > >
> > > i've been spying on this list for a few weeks now. lots of talk about
> > > foundry and alteon and cisco and such.
> > >
> > > anyone use software-based/os-level balancing instead of hardware
> > > appliances? thoughts?
> > Well, some basic drawbacks of software/os level load balancing.
> > Each server must now do additional work to make load balancing happen.
> > The load balancing code adds additional complexity to the site.
> no. there can still be machines dedicated to the balancing. the end real
> server machines need no extra complexity.
Right I was not including that (i mentioned it for resonate. I knew there
were ones for linux that did the same thing, but I didn't know the names)
> i'm still talking about network-level balancing, not "clustering" in the
> sense of sharing cpu/memory, ie: beowolf/related projects. sorry if i
> didn't clarify.
Clustering is a misused term, which is why i suggested to tony that he use
WLBS instead of windows clustering in his example.
This archive was generated by hypermail 2b30 : Thu Feb 22 2001 - 21:41:28 EST