The following message is a courtesy copy of an article
that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well.


[EMAIL PROTECTED] (Paul Gilmartin) writes:
> What does it mean to "max out"  Is it that regardless of how
> much work is available to be dispatched, the utilization
> never rises above 30%?  If so, what other bottleneck is
> constraining throughput?

sort of reminds me of when in the late 80s we were running around doing
3-tier and middle ware/layer executive presentations and taking lots of
arrows from the SAA forces. Lots of past posts
http://www.garlic.com/~lynn/subnetwork.html#3tier

part of the 3-tier and middle ware/layer executive presentations
involved e-net (as opposed to 16mbit t/r). the "word" was that e-net
only could get 1mbit thruput. However, it appeared that they must of
been using for the comparison, the original 3mbit enet before
listen-before-transmit.

there was an '88 acm sigcomm article that configuration of something
like 40 stations in low-level device driver loop constantly transmitting
minimum sized packets ... that 10mbit e-net over cat5 would drop off to
only 8mbits effective thruput.

along with that, the new almaden bldg. had been wired w/cat5 supposedly
for 16mbit t/r ... but tests were showing that 10mbit enet over the cat5
had both higher effective thruput as well as lower latency than 16mbit
t/r over the same cat5.

disclaimer: my wife had been con'ed into going to pok to be in charge of
loosely-coupled architecture. while there she came out with peer-coupled
shared data architecture ... misc past postings
http://www.garlic.com/~lynn/subtopic.html#sharedata

and also while in pok was co-inventor on token-passing ring patent (that
was subsequently granted).

and except for ims hot-standby ... the peer-couple shared data
architecture had very little uptake until sysplex.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html

Reply via email to