Heterogeneous computing scheduling with evolutionary algorithms

Abstract

This work presents sequential and parallel evolutionary algorithms (EAs) applied to the scheduling problem in heterogeneous computing environments, a NP-hard problem with capital relevance in distributed computing. These methods have been specifically designed to provide accurate and efficient solutions by using simple operators that allow them to be later extended for solving realistic problem instances arising in distributed heterogeneous computing (HC) and grid systems. The EAs were codified over MALLBA, a general-purpose library for combinatorial optimization. Efficient numerical results are reported in the experimental analysis performed on wellknown problem instances. The comparative study of scheduling methods shows that the parallel versions of the implemented evolutionary algorithms are able to achieve high problem solving efficacy, outperforming traditional scheduling heuristics and also improving over previous results already reported in the related literature.

Metaheuristics for the Virtual Machine Mapping Problem in Clouds

“…The proposed stopping criterion is an efficient execution time for on-line cloud planning, and it is in accordance with related works on grid/cloud planning in the literature (Nesmachnow et al, 2010(Nesmachnow et al, , 2012Xhafa et al, 2008).…”

Efficient Heuristics for Profit Optimization of Virtual Cloud Brokers

“…This class of methods has also often been employed in hybrid algorithms, with the objective of improving the search of metaheuristic approaches applied to solve scheduling problems in heterogeneous computing systems [26] [27].…”

Parallel metaheuristics: recent advances and new trends

“…Recently, scheduling has become a relevant problem in large modern parallel computing infrastructures such as grid, volunteer computing, and cloud computing environments. In these services, parallel metaheuristics are efficient methods to compute large scheduling in reduced execution times (Alba et al., ; Nesmachnow et al., , , ; Xhafa and Abraham, ). Software engineering and software development, a field in which parallel metaheuristics have just started to show their usefulness for exploring large search spaces, such as in the optimization of dynamic data types and memory managers in embedded systems (Risco‐Martin et al., , ), and in software testing (Alba and Chicano, ). Telecommunications, a field that has grown at a fast pace in recent years, posing difficult challenges to the research community due to the large size of the infrastructures, the need for real‐time results, etc. Parallel metaheuristics have shown a great impact on addressing these challenges by providing accurate and efficient solutions to the related optimization problems in network design (Alba and Chicano, ; Nesmachnow et al., ; Pedemonte and Cancela, ; Ribeiro and Rosseti, ), network routing (Durillo et al., ; Liu et al., ; Segura et al., ; Zhu et al., ), and network planning, especially in modern networks technologies such as cellular (Alba and Chicano, ; Talbi et al., ), mobile ad‐hoc networks (Liu et al., ), vehicular networks, sensor networks, and peer‐to‐peer (Luna et al., ; Nesmachnow et al., ).…”