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

Abstract

In Cloud Computing, designing an efficient workflow scheduling algorithm is considered as a main goal. Load balancing is one of the most sophisticated methodologies, which can optimize workflow scheduling by distributing the load evenly among available resources. A well-designed load balancing algorithm has significant impact on performance and output in Cloud Computing. Therefore, designing robust load balancing techniques to manage the networks' load has always been a priority. Researchers have proposed and examined different load balancing methods; there is, however, a large knowledge gap in adopting an efficient load balancing algorithm in the Cloud system. This paper describes how a generalized spring tensor, an evolutionary algorithm with mathematical apparatus, can be utilized for a more efficient and effective load management in Cloud Computing. Considering the fluctuation and magnitude of the load, a novel application of workflow scheduling is investigated in the context of various mathematical patterns. The preliminary results of the research show that defining the dependency ratio between workflow tasks in Cloud Computing, results in better resource management, maximized performance and minimized response time while dealing with customer's requests.

Generalized Spring Tensor Model: A New Improved Load Balancing Method in Cloud Computing

“…{ ∑ ANM, however, was suggested as a coarse gained model which is using the simpler Hookian potentials to bypass the energy minimization needed for measuring the direction of the load [9]. ANM is using Hessian matrixes shown in equation 2 with super elements, where each element is a tensor and is the interaction tensor between and [10][11].…”