Task allocation in mesh structure: 2side leapfrog algorithm and q-learning based algorithm

Authors:
Iwona Pozniak-Koszalka;Wojciech Proma;Leszek Koszalka;Maciej Pol;Andrzej Kasprzak
Affiliations:
Dept. of Systems and Computer Networks, Wroclaw University of Technology, Wroclaw, Poland;Dept. of Systems and Computer Networks, Wroclaw University of Technology, Wroclaw, Poland;Dept. of Systems and Computer Networks, Wroclaw University of Technology, Wroclaw, Poland;Dept. of Systems and Computer Networks, Wroclaw University of Technology, Wroclaw, Poland;Dept. of Systems and Computer Networks, Wroclaw University of Technology, Wroclaw, Poland
Venue:
ICCSA'12 Proceedings of the 12th international conference on Computational Science and Its Applications - Volume Part IV
Year:
2012

Citing 8
Cited 0

Technical Note: \cal Q-Learning

Machine Learning
An Efficient Task Allocation Scheme for 2D Mesh Architectures

IEEE Transactions on Parallel and Distributed Systems
A Fast and Efficient Processor Allocation Scheme for Mesh-Connected Multicomputers

IEEE Transactions on Computers
Processor Allocation in the Mesh Multiprocessors Using the Leapfrog Method

IEEE Transactions on Parallel and Distributed Systems
Design and performance evaluation of combined first-fit task allocation and migration strategies in mesh multiprocessor systems

Parallel Computing
A fast and efficient strategy for submesh allocation in mesh-connected parallel computers

SPDP '93 Proceedings of the 1993 5th IEEE Symposium on Parallel and Distributed Processing
Comparison of allocation algorithms for mesh structured networks with using multistage simulation

ICCSA'06 Proceedings of the 2006 international conference on Computational Science and Its Applications - Volume Part V
Simulation-based evaluation of distributed mesh allocation algorithms

ISPA'07 Proceedings of the 2007 international conference on Frontiers of High Performance Computing and Networking

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper concerns the problem of task allocation on the mesh structure of processors. Two created algorithms: 2 Side LeapFrog and Q-learning Based Algorithm are presented. These algorithms are evaluated and compared to known task allocation algorithms. To measure the algorithms' efficiency we introduced our own evaluating function --- the average network load. Finally, we implemented an experimentation system to test these algorithms on different sets of tasks to allocate. In this paper, there is a short analysis of series of experiments conducted on three different categories of task sets: small tasks, mixed tasks and large tasks. The results of investigations confirm that the created algorithms seem to be very promising.