Data flow computers promise efficient parallel computation limited in speed only by data dependencies in the calculation being performed. At the Massachusetts Institute of Technology Laboratory for Computer Science, the Computation Structures Group is working to design practical data flow computers that can outperform conventional supercomputers. Since data flow computers differ radically in structure from conventional (sequential) computers, the projection of their performance must be done through analysis of specific computations. The performance improvement that data flow computers offer is shown for a NASA benchmark program that implements a global weather model. We present the structure of the weather code as expressed in VAL, a functional programming language designed by the Computation Structures Group, and develop the corresponding machine-level program structures for efficient execution on a data flow supercomputer. On the basis of this analysis, we specify the capacities of hardware units and the number of each type of unit required to achieve a twenty-fold improvement in performance for the weather simulation application.
[1]
Guang R. Gao,et al.
Maximum Pipelining of Array Operations on Static Data Flow Machine
,
1983,
ICPP.
[2]
Ian Watson,et al.
A Practical Data Flow Computer
,
1982,
Computer.
[3]
WILLIAM B. ACKERMAN.
Data flow languages
,
1979,
1979 International Workshop on Managing Requirements Knowledge (MARK).
[4]
K. W. Todd.
HIGH LEVEL VAL CONSTRUCTS IN A STATIC DATA FLOW MACHINE
,
1981
.
[5]
Milton Halem,et al.
The GISS Model of the Global Atmosphere
,
1974
.
[6]
Jack B. Dennis,et al.
Building blocks for data flow prototypes
,
1980,
ISCA '80.
[7]
J. E. Rodrigues,et al.
A GRAPH MODEL FOR PARALLEL COMPUTATIONS
,
1969
.
[8]
A. L. Davis,et al.
The architecture and system method of DDM1: A recursively structured Data Driven Machine
,
1978,
ISCA '78.
[9]
Jack B. Dennis,et al.
Data Flow Supercomputers
,
1980,
Computer.
[10]
A. Arakawa,et al.
Numerical methods used in atmospheric models
,
1976
.