[MUD-Dev] Re: Question regarding Java threads

[J C Lawrence]

On Wed, 5 Aug 1998 23:23:24 -5 
Jon A Lambert<jlsysinc at ix.netcom.com> wrote:

> From: JA3:

> By the way, your P2-233 will choke with 200 active users.  Interrupt
> overflow from the NIC.  You could solve this problem with a $500 NIC
> that doesn't interrupt the CPU as much(has its own processor,) but
> then you'll have to ban char mode telnet outright and do some other
> unfriendly things too, and I just don't know that you're going to
> find 200 simultaneous players anyway; few muds do, so unless you
> become one of them, it isn't worth the time, money, and effort to
> prepare for:)

> -- John J. Adelsberger III jja at umr.edu

I can't comment on his numbers, tho 200 seems a bit low, tho its
definitely a high if you are talking X terminals and 10Mbs ethernet (X
overhead).  In very rough essence his contention is correct.  Cheap
NICs (eg NE2000) load the host CPU, using it instead of their own
non-existent intelligence.  More expensive NIC's support full duplex,
contain extensive intelligence, large caches, do direct
DMA/bus-mastering etc transfers etc, offloading work from the CPU and
the host system in general.

Core questions:  

  Can a single NIC in a single machine saturate its network (optimal
bandwidth)?  Why not?  Can a single NIC in a single machine process
all the traffic from a saturated net (say 75% for ethernet with all
traffic targetted at the same machine (promiscuous mode Okay))?  How
close can each of them get?  How much host CPU etc are required for
the attempts?

You get what you pay for.  I like 3Com and Tulip based NICs.

--
J C Lawrence                               Internet: claw at null.net
(Contractor)                               Internet: coder at ibm.net
---------(*)                     Internet: claw at under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...