No abstract

No abstract

No abstract

Some applications with gather / scatter operations are difficult to accelerate. These operations cause inefficient cache use in each processor and fine grain global communications in parallel systems. There are several applications with such characteristics particularly in electrical engineering. For examples, circuit simulation and power flow simulation with LU decomposition for random sparse matrix has such characteristics. This paper presents how to make inexpensive personal supercomputers to solve these problems. In order to get the merit of commercial-off-the-shelf (COTS) continuously after the death of vector supercomputer vendors, it is designed without any modification on CPU, bridge chips on motherboard and memory chips. Only plugging a new memory module with vector load / store function and communication functions make an inexpensive home-use personal computer into a node similar to Earth simulator's one. Applications with unit striding or indexed accesses are going to be accelerated. How to accelerate NAS CG is shown as an example.

A b t ractThis paper proposes a new programmable controller baaed on the usage of Petri Net to describe the status of a whole syrtem. The proposed programmable controller consists of Fire Unit and Input Handling Unit. And it does not UM MY processon, because pluea u p r e~ the status of the controller, not proasses. Fin Unit L an ASIC memory with very wide word to check whether a transition is fireable or not in two memory cycles. Since Input Handling Unit can choose the earliest arrival signal input using mutual exclusion rule, it is not necessary to scan d input. It takes only Cour memory cyclu, Cor the newly proposed system to process one input signal. So it is possible to realize much Cuter control that the conventional system and it has mathematically analyzable foundation.