I'm no expert here, but from my experience...
Alot of this depends on the forwarding
methoding/switching method (whatever x vendor calls
it) of a device as well as the end station. The end
stations are responsible for pieceing the packets back
together in the right order (going over multiple paths
you can get out of order packets).
Each vendor has different ways of forwarding across
equal paths. Theres per-packet, fast-switching
(building a cache based on src/dest hash), CEF, etc.
These have an effect how load balancing works as well.
On 3Com, they have 2 options for Equal Cost.
Basically, either do per-packet among the equal paths
or go over one path. BayRS has 2-3 options but I
haven't really played with it much to comment on how
well it performs/behaves.
I've worked on several networks using equal cost (OSPF
mainly) and haven't seen many issues but have seen 1-2
cases where some applications didn't work well. This
was heavy traffic as well.
I'm not familiar with F5 or Windows NT. Now, some of
those boxes do a different form of load balancing at
the upper layers to balance web-server traffic among a
group of servers.
Hope this helps.
--- Billy Monroe <[EMAIL PROTECTED]> wrote:
> How effective is the load balance features of
> routing protocols (for
> example, OSPF equal paths load balancing) when
> compared to solutions such as F5 or Windows NT load
> balancing ?
__________________________________________________
Do You Yahoo!?
Yahoo! Shopping - Thousands of Stores. Millions of Products.
http://shopping.yahoo.com/
_________________________________
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
