Master-Slave Parallel Genetic Algorithm Based on MapReduce Using Cloud Computing

2011 ◽  
Vol 121-126 ◽  
pp. 4023-4027 ◽  
Author(s):  
Guang Ming Li ◽  
Wen Hua Zeng ◽  
Jian Feng Zhao ◽  
Min Liu
Keyword(s):  
Genetic Algorithm ◽  
Cloud Computing ◽  
High Performance ◽  
Low Cost ◽  
Parallel Genetic Algorithm ◽  
Parallel Genetic Algorithms ◽  
Computing Platform ◽  
Speed Up ◽  
Cloud Computing Platform ◽  
High Performance Computer

The implementation platforms of parallel genetic algorithms (PGAs) include high performance computer, cluster and Grid. Contrast with the traditional platform, a Master-slave PGA based on MapReduce (MMRPGA) of cloud computing platform was proposed. Cloud computing is a new computer platform, suites for larger-scale computing and is low cost. At first, describes the design of MMRPGA, in which the whole evolution is controlled by Master and the fitness computing is assigned to Slaves; then deduces the theoretical speed-up of MMRPGA; at last, implements MMRPGA on Hadoop and compares the speed-up with traditional genetic algorithm, the experiment result shows MMRPGA can achieve slightly lower linear speed-up with Mapper’s number.

scholarly journals Emergency Scheduling Optimization Simulation of Cloud Computing Platform Network Public Resources

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Dingrong Liu ◽  
Zhigang Yao ◽  
Liukui Chen
Keyword(s):  
Genetic Algorithm ◽  
Cloud Computing ◽  
Objective Function ◽  
Hamming Distance ◽  
Fitness Function ◽  
Initial Population ◽  
Computing Platform ◽  
Public Resource ◽  
Public Resources ◽  
Cloud Computing Platform

Emergency scheduling of public resources on the cloud computing platform network can effectively improve the network emergency rescue capability of the cloud computing platform. To schedule the network common resources, it is necessary to generate the initial population through the Hamming distance constraint and improve the objective function as the fitness function to complete the emergency scheduling of the network common resources. The traditional method, from the perspective of public resource fairness and priority mapping, uses incremental optimization algorithm to realize emergency scheduling of public resources, neglecting the improvement process of the objective function, which leads to unsatisfactory scheduling effect. An emergency scheduling method of cloud computing platform network public resources based on genetic algorithm is proposed. With emergency public resource scheduling time cost and transportation cost minimizing target, initial population by Hamming distance constraints, emergency scheduling model, and the corresponding objective function improvement as the fitness function, the genetic algorithm to individual selection and crossover and mutation probability were optimized and complete the public emergency resources scheduling. Experimental results show that the proposed method can effectively improve the efficiency of emergency resource scheduling, and the reliability of emergency scheduling is better.

Power Management of Virtualized Cloud Computing Platform

2012 ◽  
Vol 35 (6) ◽  
pp. 1262 ◽  
Author(s):  
Ke-Jiang YE ◽  
Zhao-Hui WU ◽  
Xiao-Hong JIANG ◽  
Qin-Ming HE
Keyword(s):  
Cloud Computing ◽  
Power Management ◽  
Computing Platform ◽  
Cloud Computing Platform

Predicting Node Failures in an Ultra-Large-Scale Cloud Computing Platform

2020 ◽  
Vol 29 (2) ◽  
pp. 1-24
Author(s):  
Yangguang Li ◽  
Zhen Ming (Jack) Jiang ◽  
Heng Li ◽  
Ahmed E. Hassan ◽  
Cheng He ◽  
...  
Keyword(s):  
Cloud Computing ◽  
Large Scale ◽  
Computing Platform ◽  
Cloud Computing Platform ◽  
Node Failures

In defense of decentralized research data management

Neuroforum ◽  
2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Michael Hanke ◽  
Franco Pestilli ◽  
Adina S. Wagner ◽  
Christopher J. Markiewicz ◽  
Jean-Baptiste Poline ◽  
...  
Keyword(s):  
Cloud Computing ◽  
Data Management ◽  
Research Data ◽  
Data Archive ◽  
Individual Scientist ◽  
Computing Platform ◽  
Research Data Management ◽  
Cloud Computing Platform ◽  
Research Objects ◽  
Central Services

Abstract Decentralized research data management (dRDM) systems handle digital research objects across participating nodes without critically relying on central services. We present four perspectives in defense of dRDM, illustrating that, in contrast to centralized or federated research data management solutions, a dRDM system based on heterogeneous but interoperable components can offer a sustainable, resilient, inclusive, and adaptive infrastructure for scientific stakeholders: An individual scientist or laboratory, a research institute, a domain data archive or cloud computing platform, and a collaborative multisite consortium. All perspectives share the use of a common, self-contained, portable data structure as an abstraction from current technology and service choices. In conjunction, the four perspectives review how varying requirements of independent scientific stakeholders can be addressed by a scalable, uniform dRDM solution and present a working system as an exemplary implementation.

Sign in / Sign up

Export Citation Format

Share Document