系統解析計算における粒度と並列計算の効率およびアプリケーションの形式

抄録

Much study to enhance the calculation speed of power system analysis by means of utilizing parallel systems have been proposed. However, the speedup gained in these systems have been small compared to other applications such as circuit simulation mainly due to the smallness of the problem size of power calculations. The before mentioned studies had concentrated on so called “space parallel computing”, therefore parallel implementation grain is small due to the inherent size of power system calculation.
In this paper, we first study the effect that the ratio of data transmission time to calculation time per grain has on the speedup of total calculation. We show that the smaller ratio raises the efficiency of parallel implementation. Lasger grain size of the problem generally yields smaller ratio. Thus, it is shown that attaining a larger grain size is important. In the power system calculation, there is a method called time-domain-parallel computing whose algorithm produces larger grain size. The other application form which results in much larger grain size is multiple cases analysis. We studied the composition methods of these parallel implementations of transient stability analysis using a MIMD distributed memory type parallel processor system.
On a 16 PE system, we attained a speedup of 6.3 in time-domain-parallel computing and a speedup of up to 15.8 in multiple cases parallel analysis. These values are much larger than the speed up in space parallel computing which was less than 2.0