A quantum‐inspired binary gravitational search algorithm–based job‐scheduling model for mobile computational grid

Abstract

Owing to the advancements in low-power consumption processors and high-power storage in a small-sized battery, the cost of handheld mobile devices, eg, mobiles, tabs, or personal digital assistants, have reduced to a great extent. This has enabled people to have at least 1 smartphone in general with this number increasing exponentially. However, the increasing use of these mobile devices results in an equal increase in the underused processing capacity of these devices too. This encourages the research aiming to use this processing power by forming a mobile computational grid. Because of the inherent limitations of bandwidth, battery, and computational power, job scheduling on these devices demands an efficient scheduling approach to harness the true potential of the grid. The problem becomes even more challenging considering the dynamic nature of these mobile devices. Job scheduling being nondeterministic polynomial time-complete allows the use of evolutionary approaches by exploring and exploiting the search space efficiently. The exploration gets boosted even more with the use of quantum-computing concepts. This work proposes a quantum-inspired Newtonian approach of attraction based on gravitational search algorithm for scheduling the jobs on mobile computational grid. Simulation study has been performed to evaluate the performance of the model over various dimensions. A comparative study has been performed with quantum-genetic algorithm. Simulation result establishes the effectiveness of model under various test conditions. KEYWORDS genetic algorithm (GA), gravitational search algorithm (GSA), mobile grid computing (MGC), quantum computing (QC), quantum-inspired genetic algorithm (QGA), turnaround time (TAT)

DewSim: A trace‐driven toolkit for simulating mobile device clusters in Dew computing environments

“…Most of the advances in the area were performed with in‐house simulation development efforts that, when not attending publication specific goals, are hard to reuse because source code is not publicly available and/or employed techniques are not properly documented. Within these efforts are those built on top of generic simulation libraries, eg, JAVASim, or those that utilize mathematical and statistical functions of MATLAB . Others adapted specific simulation frameworks such as OMNET++, NS‐2, and SimGrid, which were initially designed for different resource management purposes.…”

“…Within these efforts are those built on top of generic simulation libraries, eg, JAVASim, 24 or those that utilize mathematical and statistical functions of MATLAB. 5,25 Others adapted specific simulation frameworks such as OMNET++, 8 NS-2, 26 and SimGrid, 27 which were initially designed for different resource management purposes. It means that simulation setup requires from modelers to learn and instantiate objects and relations that deviate from the main research domain.…”

“…The names relate, to some extent, with the computing paradigm under which such exploitation is contextualized. For instance, mobile Grid computing stems from the fact that groups of SDs operate as computing providers in Grid computing infrastructures. Local mobile clouds or mobile device clouds are proposed as an alternative to execute workload when the access to a Cloud infrastructure is not available or prohibitive.…”

Towards Integrating Mobile Devices into Dew Computing: A Model for Hour-Wise Prediction of Energy Availability

“…One of these is the devices mobility perspective where can be found works focused on designing indicators to quantify the time a device is reachable by a central controller node from where tasks are distributed. For example, in [21] historical information of movements through different wireless coverage regions visited by a mobile device user, in working hours during a week, are summarized into a mobility score. The score is used by the proposed quantum-inspired binary gravitational search scheduler to rank devices according to their expected physical presence at a point of interest.…”

A Dynamic Quantum-Inspired Genetic Algorithm with Lengthening Chromosome Size

“…In [4][5][6] quantum-inspired metaheuristics are applied to image processing problems, structural design applications are presented in [7][8][9][10]. In [11][12][13][14][15][16], quantum-inspired metaheuristics are utilized for job-scheduling problem, and some examples of networks applications of quantum-inspired metaheuristics are found in [17][18][19][20]. Other applications are feature selection [21?…”