Scheduling in Heterogeneous Computing and Grid Environments Using a Parallel CHC Evolutionary Algorithm

Abstract

Scheduling is a capital problem when using distributed heterogeneous computing (HC) and grid environments to solve complex problems. The scheduling problem in heterogeneous environments is NP-hard, so a significant effort has been made to develop efficient methods for solving the problem. However, few works have faced realistic grid-sized problem instances. This work presents a parallel CHC (pCHC) evolutionary algorithm codified over MALLBA, a general-purpose library for combinatorial optimization, for solving the scheduling problem in HC and grid environments. Efficient numerical results are reported in the experimental analysis performed on both a standard benchmark and a set of large-sized problem instances specially designed in this work. The comparative study shows that pCHC is able to achieve high problem solving efficacy, significantly improving over traditional deterministic scheduling methods, while also showing a good scalability behavior when solving large problem instances.

Multiobjective Variable Neighborhood Search algorithm for scheduling independent jobs on computational grid

“…The results revealed that GA achieved the best results compared with the batch queuing heuristics. Some of the job scheduling algorithms are nature-inspired, e.g., SA [17], Ant Colony Optimization [18], Particle Swarm Optimization [19], Differential Evolution (DE) [20], parallel Cross generational elitist selection, Heterogeneous recombination, and Cataclysmic mutation (pCHC) [21]. There are also non-nature-inspired metaheuristics, such as TS [22], Threshold Accepting (TA) [23], and VNS algorithm [24].…”

“…This paper considers the test instances proposed by Nesmachnow et al, with dimension 1024 · 32, 2048 · 64, and 4096 · 128 [21]. The grid job scheduling algorithm was developed using MATLAB R2010a and run on an Intel(R) Core(TM) i5 2.67 GHz CPU with 4 GB RAM.…”

“…Nesmachnow et al [21] proposed a test suite of several large dimension heterogeneous computing scheduling problem instances, in order to model large heterogeneous computing clusters and medium sized grid infrastructures. All test instances of each dimension are composed of m grid nodes and n jobs, which is referred as the configuration m · n.…”

“…The concept of static scheduling frequently appears in many scientific research problems, especially in Single-Program Multiple-Data applications used for multimedia processing, scientific computing, data mining, parallel domain decomposition of numerical models for physical phenomena, etc. The independent tasks model also arises when different users submit their tasks to execute in volunteer-based and grid computing services and in parameter sweep applications, which are structured as a set of multiple experiments, each one executed with a different set of parameter values [21].…”

Group-based Parallel Multi-scheduler for Grid computing

“…The experimental evaluation of the proposed parallel methods demonstrates that a significant reduction on the computing times can be obtained when using the parallel GPU hardware. Further approaches have proposed evolutionary algorithms which exploit GPUs in solving the scheduling problem [18,19]. Nesmachnow, Cancela and Alba [20] implemented a parallel micro evolutionary algorithm to schedule tasks in heterogeneous and Grid environment algorithm.…”