Messages in this thread | | | Subject | Re: Why does Mindcraft insist on 4* 100BaseTX? | Date | Sun, 9 May 1999 15:22:58 +0300 (EEST) | From | Matti Aarnio <> |
| |
Alex Belits <abelits@phobos.illtel.denver.co.us> replied to mr. Spraul: > On Sun, 9 May 1999, Manfred Spraul wrote: > > I've read the Mindcraft Open benchmark invitation, > > and Mindcraft (ie Microsoft) require 4*100BaseTX. > > > > It that a common installation, or do you usually > > use Gigabyte ethernet? .... > This configuration is necessary to create high enough load with clients > that don't have gigabyte ethernet.
No, more likely having "big" cisco switch with Giga-Ether port(s) costs so much that they didn't even think of it.. Check what cisco WS-C2948G costs to you. 2x GE ports, 48x 10/100 BaseTX ports. (No need to get those high-end WS-C5***, WS-C6***, or WS-C85** series boxes.)
Of course they might not have had a visit from a marketroid over last 3 months what that C2948G has been available, so perhaps they just don't know of them...
With that kind of "middle-hardware" it does not matter that the clients don't have GE ports.
Of course in real life I would use Ether Channel bonded group of interfaces if I have a *large* number of clients, or more likely just simply multiple interfaces with separate addresses, and DNS round-robin (or manual setup) to give out those addresses to the clients to distribute the load evenly among the interfaces. (The Ether Channel does use *one* interface towards given MAC address, so doing balancing by using separately addressed inter- faces gives the same result -- possibly even better when making balancing manually controlled, and thus possibly perfectly evenly distributed.)
> -- > Alex
/Matti Aarnio <matti.aarnio@sonera.fi>
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |