QoS Constrained Grid Workflow Scheduling Optimization Based on a Novel PSO Algorithm

Abstract

Currently, computational grid, service grid and data grid are becoming richer and more complex. Grid infrastructure composed of heterogeneous resource is widely distributed, workflow scheduling problem in grid environment described by directed acrylic graph (DAG) becomes an important and difficult problem. Meanwhile, the research on workflow scheduling problem in grid environment mainly focuses on the time and cost constrained optimization, but the key problems about stability, flexibility, security and load balancing aren't considered adequately. Aiming at these problems, we redefine parameters of quality of service (QoS) and the model of grid workflow scheduling, and put forward a rotary hybrid discrete particle swarm optimization (RHDPSO) algorithm, in which double extremums are disturbed by the method of random time sequence based on rotation discretization, to overcome premature convergence and local optimum. The simulation results show that the RHDPSO algorithm has fast convergence, high precision and strong robustness, and can effectively restrain premature convergence, compared with DPSO. The performance of our algorithm is very promising.

Cost-aware challenges for workflow scheduling approaches in cloud computing environments: Taxonomy and opportunities

“…data storage cost, resource cost, resource computation cost, I/O cost, resource wastage, required network resources, service provider cost, etc.) [12,13,52,65,89]. It is possible that the resource request to begin earlier than the projected start time.…”

“…Notice that by ensuring the security of potential resources certainly decreases the possibility of being attacked and damaged to a great extent [89]. Similarly, security is considered as an important quality aspect of the cloud service that ensures their confidentiality by authenticating the parties involved, encrypting messages, and providing access control [49,79,90].…”

A Survey: Particle Swarm Optimization-Based Algorithms for Grid Computing Scheduling Systems

“…The drawback of this proposal is that the hillclimbing heuristic scans each particle which renders local search computation complex (Engelbrecht, 2005). Tao et al (2009) proposed RHDPSO and RCPSO (Tao et al, 2011b) to solve multi-QoS constrained grid workflow scheduling problem. PSO particles represent DAG task graphs; dimensions of a particle represent the tasks in the corresponding graph.…”

Generalized Spring Tensor Algorithms: with Workflow Scheduling Applications in Cloud Computing

“…Tao [7] and Suraj [8] designed a workflow load balancing algorithm based on particle swarm optimization. Considering the robustness characteristic, the algorithm mainly focuses on costs minimizations.…”