Written on cydrome label on bottom frame inside the back door od the large cabinet.
72-033174-001 REV A
Written on cydrome label on bottom frame inside back door of small cabinet.
72-201000-101 REV A2
Written on cydrome label on bottom frame inside the front door of the large cabinet.
"System identification no." Typed on a cydrome label on the outside of the rear door.
overall: 66 1/2 in x 89 1/2 in x 35 in
Donor writes [8/13/93]: "The Cydra 5 is a multi-processor, with a single VLIW-style Numeric Processor (NP) for numerical applications, plus multiple scalar processors for I/O and other general-purpose, front-end processing. The NP has a 256-bit instruction word capable of issuing 7 operation per cycle. The NP architecture is unique in its support of executing all manner of loops (vectorizable or otherwise) with maximum parallelism. Amongst other features, it is the first product to provide full-blown predicated (guarded) execution, rotating registers plus other support for software pipelining. The Cydra 5's memory system too is unique in that it is the first product implementation of an interleaved memory system which is insensitive to the stride with which an array is being accessed. This is accomplished by the use of a novel pseudo-random hashing scheme. The machine is more fully documented in the January issue of Computer magazine and in the Journal of Supercomputing, vol. 7, no. 1/2." From , Cydrome Cydra 5, 1988 - Bob Rau, Ross Towle, David Yen, and Wei Yen Ross Towle - compiler worked for Gary Beck on the processor: Joe Bratt, John Brennan, Steve Wilson, and Gulbin Ezer worked on the memory system - Ed Wolff and Norman Yeung worked for Jack Kister on the I/O system: Gil Lauer, Jack Mills, David Roe, and Mohammed Seth others who contributed to hardware and software efforts: Jim Dehnert, Stimson Ho, and Mike McNammara Peter Hsu and Mike Schlansker did the groundwork for the Cydra 10 (which did not get further than a paper design before the company folded) Peter Hsu went to MIPS and the MIPS R8000 was greatly influenced by the Cydra designs See Bob Rau, David Yen, Wei Yen, and Ross Towle, "The Cydra 5 Departmental Supercomputer: Design Philosophies, Decisions, and Trade-offs," IEEE Computer, January 1989, pp. 12-35. Letters Regarding Transfer in Object File