130 likes | 294 Views
MRPGA : An Extension of MapReduce for Parallelizing Genetic Algorithm. Reporter :古乃卉. Outline. Abstract Introduction Related Work Architecture MRPGA Implementation Experiments Conclusion. Abstract. MapReduce Map and Reduce Genetic Algorithm Iteration MRPGA
E N D
MRPGA:An Extension of MapReduce for Parallelizing Genetic Algorithm Reporter:古乃卉
Outline • Abstract • Introduction • Related Work • Architecture • MRPGA • Implementation • Experiments • Conclusion
Abstract • MapReduce • Map and Reduce • Genetic Algorithm • Iteration • MRPGA • Extension of MapReduce for Parallelizing Genetic Algorithm
Introduction • Problems of Parallelized Genetic Algorithm • Communication, synchronization, heterogeneity and frequent failures • Why MapReduce? • Provides a parallel design pattern for simplify application developments • How to work? • Add a phase for global selection at the end of every iteration of PGAs and a coordinator
Related Work • PGAs • Distributed, coarse grained and fine grained • MPI:not flexible enough for handling heterogeneity and failures • MapReduce • Phoenix, Hadoop and MRPSO
MRPGA • Map, Reduce and Reduce • Key:index of the individual • Value:the individual • Allows each of the reduce tasks to collect dependent input without fetching data from a remote machine
MRPGA(cont.) • Key:individual • Value:just number
MRPGA(cont.) • Select the global • Optimum individual • Reproduction, mutation and submission of offspring to the scheduler of MRPGA , and collection optimum individual
Experiments • MRPGA runtime system with Aneka • An enterprise Grid consisting of 33 nodes • Pentium 4 processor • 1GB of memory • 160GB IDE disk • 1 Gbps Ethernet • Windows XP
Experiments(cont.) • 300 individuals • 100 generations • Simulated cost • Avg. evaluation 10 sec. • Standard deviation 0.2 • 500 individuals • 10 times MOAE MOAE+MRPGA
Conclusion • This extension makes PGAs can benefit from the MapReduce model on handling heterogeneity and failures