PACKING SEQUENTIAL STRETCHES IN THE MDFM

The Manchester Dataflow Machine (MDFM) works with tasks of size equal to one single instruction. This fine granularity aims at exploring all parallelism at the instruction level. However, this project decision increases the instruction communication cost, which ends up to jam the interconnection network and reduces the system performance. One way t o skirt this problem is to adopt variable size tasks instead of workin g with such small task size. In this paper, in order to study whether or not the usage of such variable size tasks in the MDFM architecture contributes to the improvement of the performance, some simulations by toy programs take place. In the simulation, variable size tasks are r ealized by packing the sequential instruction stretches into one task. To manage this packing, the Sequential Block (SE) technique is develo ped. The simulation of those packed and unpacked programs give an outl ine of advantages and disadvantages of working with variable size task s, and how the SE technique should be implemented in the system.