The Supercomputer Toolkit is a family of hardware and software modules from which high-performance special-purpose computers for scientific/engineering use can be easily constructed and programmed. The hardware modules include processors, memory, I/O devices and communication devices. The software modules include an operating system, compilers, debuggers, simulators, scientific libraries, and high-level front ends. When faced with a suitable problem, the engineer/scientist connects the modules by means of static-interconnect technology and constructs a problem-specific parallel computation network. The network is loaded from a workstation that serves as a host. When the program is run, results are collected and displayed by the host. The host handles files, does compilation, etc. The computation network, the Toolkit, does the heavy computation. In addition to high performance, the advantage of the Toolkit is its low cost which makes it potentially affordable by small groups as their main number crunching computer. This paper is concerned with the Toolkit version built at the Technion, which is a second generation of the MIT version.1 The paper briefly describes the hardware and software of this new version and its application to elastic-plastic flow, weather prediction and the simulation of electronic circuits. The main topic of the application section is the relation between the Toolkit configuration and the computation structure of these applications. The paper discusses conclusions related to the hardware and software as well as to the techniques for applying the system.
[1]
B. J. Adler.
Special purpose computers
,
1988
.
[2]
Harold Abelson,et al.
The Supercomputer Toolkit: A General Framework for Special- purpose Computing
,
1991
.
[3]
Harold Abelson,et al.
The Dynamicist''s Workbench: I ---Automatic Preparation of Numerical Experiments
,
1987
.
[4]
Andrew A. Berlin,et al.
Partial evaluation applied to numerical computation
,
1990,
LISP and Functional Programming.
[5]
R A Kerr.
From mercury to pluto, chaos pervades the solar system.
,
1992,
Science.
[6]
Shekhar Y. Borkar,et al.
iWarp: an integrated solution to high-speed parallel computing
,
1988,
Proceedings. SUPERCOMPUTING '88.
[7]
G. Sussman,et al.
Numerical Evidence That the Motion of Pluto Is Chaotic
,
1988,
Science.
[8]
Andrew A. Berlin,et al.
Compiling scientific code using partial evaluation
,
1990,
Computer.
[9]
G. Sussman,et al.
Chaotic Evolution of the Solar System
,
1992,
Science.