Reliability oriented task allocation in heterogeneous distributed computing systems

• Published in: ISCC 2004 : Ninth IEEE International Symposium on Computers and Communication : proceedings : June 28-July 1, 2004, Alexandria, Egypt Vol. 1; pp. 68 - 73 Vol.1
• Main Authors: Attiya, G., Hamam, Y.
• Format: Conference Proceeding
• Language: English
• Published: Los Alamitos CA IEEE 2004; IEEE Computer Society
• Subjects: Application software; Applied sciences; Computer networks; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Cost function; Distributed computing; Exact sciences and technology; Hardware; Operational research and scientific management; Operational research. Management science; Physics computing; Redundancy; Reliability; Reliability theory. Replacement problems; Resource management; Software; Workstations; Optimal algorithm; Branch and bound method; Task scheduling; Distributed system; System reliability; Computational complexity; Modeling; Multiagent system; Optimal allocation; Algorithm complexity; Heuristic method; Simulated annealing; Time complexity
• ISBN: 9780780386235; 078038623X
• DOI: 10.1109/ISCC.2004.1358383

This work tackles the problem of task allocation in heterogeneous distributed computing systems with the goal of maximizing the system reliability. It first develops a mathematical model for reliability based on a cost function representing the unreliability caused by the execution of tasks on the system processors and the unreliability caused by the interprocessor communication costs subject to constraints imposed by both the application and the system resources. It then proposes an optimal, memory efficient, hybrid algorithm to this problem. The algorithm first finds a near optimal allocation by applying the well known simulated annealing (SA) and then finds an optimal distribution by applying the branch-and-bound (BB) technique considering the solution of the SA as the initial solution. The proposed algorithm overcomes the low solutions quality that may be obtained by using heuristics. It also overcomes the computational time complexity of the exact algorithms.