🚀 go-pugleaf

In article <9600@june.cs.washington.edu> david@june.cs.washington.edu.cs.washington.edu (David Callahan) writes:
>In article <224@dg.dg.com> uunet!dg!daver (David Rudolph) writes:
>>In article <36662@lll-winken.LLNL.GOV> brooks@maddog.llnl.gov (Eugene Brooks) writes:
>>>In article <20764@usc.edu> vorbrueg@bufo.usc.edu (Jan Vorbrueggen) writes:
>>>>Message forwarding isn't so difficult either. I've read of a system
>>>>requiring less than 10 us overhead per through-route
>>>This level of overhead for each hop is completely intolerable.
>
>>And what is the overhead for each interconnect level in a "shared
>>memory" machine such as the ultracomputer?  Keep in mind that in such a
>>machine, every non-local memory access must go through each level and 
>>back while the processor waits.

In a machine like the BBN TC-2000 with a butterfly type switch. All remote
memory is equidistant from all others. Each switch stage takes 25 nS in each 
direction, and there are at most 3 stages (in a 504 node machine).

The normal time for a 32 bit read or write is around 1.5 uS to any other
node.

The time for local references tends to be more in the 200 nS area.


Kent Hoult
TEL: (617) 873-4385     ARPA: khoult@bbn.com