Abstract
In recent years, grid and mesh structures have received increasing attention. The mesh based multicomputers are the future of processing. As we slowly reach natural limits of semi-conductor spatial density supercomputer design depends more heavily on parallel and distributed processing. This paper concerns mesh allocation algorithms effectiveness assessment and the experimentation system that was developed to provide testing environment. Most focus was put on creation of such a system that would represent the highest scope of real supercomputers inner working routines and at the same time supply a way to input measured processing data as a base of allocation algorithm load computation. In investigations reported different allocation algorithms, including own WSBA, and various task parameters are considered.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Pozniak-Koszalka, I., Koszalka, L., Kubiak, M.: Allocation Algorithm for Mesh-Structured Networks. In: Proc. of 5th ICN Conference IARIA, ICN5: Management, vol. 24, IEEE Computer Society Press, Los Alamitos (2006)
Koszalka, L., Kubiak, M., Pozniak-Koszalka, I.: Comparison of SBA-Family Task Allocation Algorithms for Mesh Structured Networks. In: ISPA Workshops, LNCS, pp. 21–30 (2006)
Chang, C., Mohapatra, P.: An Integrated Processor Management Scheme for Mesh-Connected Multicomputer System. In: Proc. of International Conference on Parallel Processing (August 1997)
Koszalka, L., Lisowski, D., Pozniak-Koszalka, I.: Comparison of Allocation Algorithms for Mesh - Structured Networks with Using Multistage Simulation. In: Gavrilova, M., Gervasi, O., Kumar, V., Tan, C.J.K., Taniar, D., Laganà , A., Mun, Y., Choo, H. (eds.) ICCSA 2006. LNCS, vol. 3984, pp. 58–67. Springer, Heidelberg (2006)
Byung, S., Das, C.R.: A Fast and Efficient Processor Allocation Scheme for Mesh-Connected Multicomputers. IEEE Trans. on Computers 1, 46–59 (2002)
Koszalka, L.: Static and Dynamic Allocation Algorithms in Mesh Structured Networks, ICDCIT, LNCS, pp. 89–101 (2006)
Sharma, D.D., Pradhan, D.K.: Submesh Allocation in Mesh Multicomputers Using Busy-List: A Best-Fit Approach with Complete Recognition Capability. Journal of Parallel and Distributed Computing 1, 106–118 (1996)
Chu, Y-K., Yen, I-L., Rover, D.T.: Guiding Processor Allocation with Estimated Execution Time for mesh Connected Multiple Processor Systems. In: Proc. of 28-th Annual HICSS, pp. 163–172 (1995)
De, M., Das, D., Ghosh, M.: An Efficient Sorting Algorithm on the Multi-Mesh Network. IEEE Trans. on Computers 10, 1132–1136 (1997)
Yang, Y., Wang, J.: Pipelined All-to-All Broadcast in All-Port Meshes and Tori. IEEE Trans. on Computers 10, 1020–1031 (2001)
Kasprzak, A.: Packet Switching Wide Area Networks, WPWR, Wroclaw, /in Polish/ (1997)
Liu, T., Huang, W., Lombardi, F., Bhutan, L.N.: A Submesh Allocation for Mesh-Connected Multiprocessor Systems. In: Parallel Processing, pp. 159–163 (1995)
Batcher, K.E.: Architecture of a massively parallel processor. In: Proc. of International Conference on Computer Architecture, pp. 174-179 (1998)
(2006), http://www.top500.org/
Jakubowsk, J., Stencel, R.: Introduction to Probability Theory, SCRIPT, /in Polish/ (2001)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Koszalka, L. (2007). Simulation-Based Evaluation of Distributed Mesh Allocation Algorithms. In: Thulasiraman, P., He, X., Xu, T.L., Denko, M.K., Thulasiram, R.K., Yang, L.T. (eds) Frontiers of High Performance Computing and Networking ISPA 2007 Workshops. ISPA 2007. Lecture Notes in Computer Science, vol 4743. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74767-3_35
Download citation
DOI: https://doi.org/10.1007/978-3-540-74767-3_35
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-74766-6
Online ISBN: 978-3-540-74767-3
eBook Packages: Computer ScienceComputer Science (R0)