[email protected] (Anne & Lynn Wheeler) writes:
> Part of the issue with serial fibre-optic for fibre channel standard
> http://en.wikipedia.org/wiki/Fibre_Channel
> and similar work going on about the same time with scalable coherent
> interface (that I also got dragged into)
> http://en.wikipedia.org/wiki/Scalable_Coherent_Interface
>
> was to trying to eliminate the increasing end-to-end latency penalty in
> protocols. The objective was to move to protocols that just treated the
> media as raw continuous transport.

re:
http://www.garlic.com/~lynn/2013m.html#94 SHARE Blog: News Flash: The Mainframe 
(Still) Isn't Dead

pure aside, in the late 80s, the IBM Hursley lab did something similar
to FCS & SCI (and what I worked with for channel extender in 1980) with
Harrier/9333 ... at the time it was 80mbit/sec serial "copper" links
(concurrent in both directions, not fiber-optic) ... they packetized
scsi i/o program and downloaded to remote end ... where it was executed
... significantly reducing the latency and significantly increasing
aggregate throughput (compared to standard scsi bus ... where actual
physical disk drives in the benchmark comparisons were essentially
otherwise identical).

as part of ha/cmp, I was hoping to see harrier/9333 evolve into
inter-operable fractional FCS ... old post with some discussion
http://www.garlic.com/~lynn/95.html#13
http://www.garlic.com/~lynn/96.html#15

however, end of Jan1992, cluster scaleup is transferred and we are told
we can't work on anything with more than four processors ... and we
decide to leave. Then Harrier/9333 evolves into (non-interoperable) SSA
with serial copper upgraded/doubled to 160mbits/sec (concurrent in both
directions) or approx. 40mbyte/sec aggregate
http://en.wikipedia.org/wiki/Serial_Storage_Architecture

by comparison ESCON is 200mbits/sec links or 400mbits/sec (50mbyte/sec)
aggregate ... but is lucky to achieve 17mbyte/sec aggregate throughput

small mention of 9333 here
http://en.wikipedia.org/wiki/History_of_IBM_magnetic_disk_drives

some old email
http://www.garlic.com/~lynn/2006p.html#email881011
http://www.garlic.com/~lynn/2009q.html#email901024
http://www.garlic.com/~lynn/2010.html#email920629
http://www.garlic.com/~lynn/2001f.html#email920630
http://www.garlic.com/~lynn/2010.html#email960203

-- 
virtualization experience starting Jan1968, online at home since Mar1970

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to