One of the hot topics in graphic processing unit (GPU) research is workload scheduling. For parallel workloads with a large scale, the scheduling strategy can affect seriously system performance. To address this, the ...
详细信息
One of the hot topics in graphic processing unit (GPU) research is workload scheduling. For parallel workloads with a large scale, the scheduling strategy can affect seriously system performance. To address this, the authors carry out scheduling of data transfer before workload execution scheduling, and propose an optimal scheduling algorithm for GPU workload. By hiding data transfer in workload execution to the maximum extent, the algorithm can reduce wait time, thus achieving a small timespan. They attribute the problem of hiding data transfer in workload execution to the 0-1 knapsack problem, and propose the pseudo-polynomial timealgorithm based on the Dyer-Zemel algorithm. The authors then deduce the fully polynomial-timealgorithm scheme for PPTA. By testing on cloud platform equipped with Nvidia Geforce GTX 750, they show that their scheduling algorithm estimates the optimal schedule sequence effectively.
We consider the problem of finding a shortest solution for the Post correspondence problem over a unary alphabet. We show that the complexity of this problem heavily depends on the representation of the input: the pro...
详细信息
We consider the problem of finding a shortest solution for the Post correspondence problem over a unary alphabet. We show that the complexity of this problem heavily depends on the representation of the input: the problem is NP-complete if the input is given in compact (logarithmic) form, whereas it becomes polynomially solvable if the input is encoded in unary. (C) 2003 Elsevier Science Ltd. All rights reserved.
暂无评论