Name
Abstract | ||
|---|---|---|
In this paper, genetic algorithm (GA) accelerated by Intel Xeon Phi coprocessor based on Intel Many Integrated Chip (MIC) Architecture is proposed and called GAPhi framework. The GAPhi framework solves the power-aware task scheduling (PATS) problems in shorter execution time than sequential genetic algorithm. We evaluate GAPhi, sequential GA (SGA) and GAGPU for solving the same problem size of PATS problems. Due to limited hardware resources (i.e. memory) for executing simulation, we created a workload that contains maximum problem size of 1000 jobs and 1000 physical machines. The experimental results show the GAPhi program executed on a single Intel Xeon Phi coprocessor (61 cores) obtains significant speedup in comparison to the SGA program executed on CPU Intel® Xeon and GAGPU program executed on NVIDIA Tesla with same input problem size. They share the same GAu0027s parameters (e.g. number of generations, crossover and mutation probability, etc.). |
Year | Venue | Field |
|---|---|---|
2017 | SoICT | Crossover,Scheduling (computing),Xeon Phi,Computer science,Parallel computing,Chip,Execution time,Xeon,Genetic algorithm,Speedup |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
2 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Nguyen Quang-Hung | 1 | 50 | 6.06 |
| Anh-Tu Ngoc Tran | 2 | 0 | 0.34 |
| Nam Thoai | 3 | 70 | 18.86 |