Basic HTML version of Foils prepared
17 November 98
Foil 39 Scalable Interconnects
From
Master Foilset for HPC Achitecture Overview CPS615 Introduction to Computational Science --
Fall Semester 1998
.
by
Geoffrey C. Fox
1
Problem is interconnect: cost (crossbar) or bandwidth (bus)
Hypercube was an attempt at compromise between these two
2
Dance-hall: bandwidth still scalable, but lower cost than crossbar
latencies to memory uniform, but uniformly large
3
Distributed memory or non-uniform memory access (NUMA)
Construct shared address space out of simple message transactions across a general-purpose network (e.g. read-request, read-response)
4
Caching shared (particularly nonlocal) data?
in Table To:
©
Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu
If you have any comments about this server, send e-mail to
webmaster@npac.syr.edu
.
Page produced by
wwwfoil
on Sun Apr 11 1999