Architectural Support for Task Migration Concerning MPSoC

• Alexandra Aguiar PUCRS
• Sérgio J. Filho PUCRS
• Tatiana G. dos Santos UNISC
• César Marcon PUCRS
• Fabiano Hessel PUCRS

Resumo

Embedded computing systems are currently present in a wide range of consumer goods and their main characteristic is the implicit specialized behavior but keeping a certain level of flexibility once it avoids redesigning due to small requirement changes. Thus, microprocessors usually are a good alternative to achieve such flexibility. Consequently, embedded systems designs make use of homogeneous or heterogeneous processors families for its complete implementation, which is known as Multiprocessor System-on-Chip (MPSoC), which can have a performance speed up through using dynamic load balancing strategies, such as task migration, to fairly distribute the existing tasks among all embedded processors. The objectives of this work are to discuss architectural aspects for embedded systems, which allow a dynamic task migration and its implications, as well as to present different techniques for the dynamic task migration showing their possible use in MPSoCs.

Referências

Abubakar, H. R. and Aftab, U. (2004). Evaluation of load balancing strategies. In Proceedings of National Conference on Emerging Technologies.

Barcelos, D., Brião, E. W., and Wagner, F. R. (2007). A hybrid memory organization to enhance task migration and dynamic task allocation in NoC-based MPSoCs. In SBCCI ’07: Proceedings of the 20th annual conference on Integrated circuits and systems design, pages 282–287, New York, NY, USA. ACM.

Bertozzi, S., Acquaviva, A., Bertozzi, D., and Poggiali, A. (2006). Supporting task migration in multi-processor systems-on-chip: A feasibility study. In Design, Automation and Test in Europe, 2006. DATE ’06. Proceedings, pages 1–6.

Brião, E., Barcelos, D., Wronski, F., and Wagner, F. R. (2007). Impact of task migration in NoC-based MPSoCs for soft real-time applications. In IFIP INTERNATIONAL CONFERENCE ON VERY LARGE SCALE INTEGRATION, New York, NY, USA.

Casavant, T. L. and Kuhl, J. G. (1988). A taxonomy of scheduling in general-purpose distributed computing systems. IEEE Trans. Softw. Eng., 14(2):141–154.

Chang, H. W. D. and Oldham, W. J. B. (1995). Dynamic task allocation models for large distributed computing systems. IEEE Trans. Parallel Distrib. Syst., 6(12):1301–1315.

Cores, O. (2007). Plasma most MIPS I(TM) opcodes. Available at [link]. Accessed at 01 dez.

Coskun, A. K., Rosing, T. S., and Whisnant, K. (2007). Temperature aware task scheduling in MPSoCs. In Design, Automation and Test in Europe Conference and Exhibition, 2007. DATE ’07, pages 1–6.

Council, J. E. D. E. (2006). Failure mechanisms and models for semiconductor devices. [link].

Dandamudi, S. P. (1998). Sensitivity evaluation of dynamic load sharing in distributed systems. IEEE Concurrency, 6(3):62–72.

de Mello, R. F. and Senger, L. J. (2006). Modelo de migração baseado na avaliação da carga e tempo de vida de processos em ambientes heterogêneos. IEEE Latin America Transactions, 4(5).

Farines, J.-M., da Silva Fraga, J., and de Oliveira, R. S. (2000). Sistemas de Tempo Real. Second Escola de Computação, IME-USP, São Paulo-SP.

Fröhlich, A. A. and Schröder-Preikschat, W. (1999). High performance application-oriented operating systems, the EPOS aproach. In Proceedings of the 11th Symposium on Computer Architecture and High Performance Computing, pages 3–9.

Götz, M., Xie, T., and Dittmann, F. (2007). Dynamic relocation of hybrid tasks: A complete design flow. In Sassatelli, G., Glesner, M., Bobda, C., and Benoit, P., editors, ReCoSoC, pages 31–38. Univ. Montpellier II.

Jerraya, A., Tenhunen, H., and Wolf, W. (2005). Multiprocessor systems-on-chips. Computer, 38(Issue 7):36–40.

Lan, Y. and Yu, T. (2001). A dynamic central scheduler load balancing mechanism. IEEE Trans. on Parallel and Distributed Systems, 12(9):899–911.

Nollet, V., Avasare, P., Mignolet, J.-Y., and Verkest, D. (2005). Low cost task migration initiation in a heterogeneous MP-SoC. In DATE ’05: Proceedings of the conference on Design, Automation and Test in Europe, pages 252–253, Washington, DC, USA. IEEE Computer Society.

Streichert, T., Strengert, C., Haubelt, C., and Teich, J. (2006). Dynamic task binding for hardware/software reconfigurable networks. In SBCCI ’06: Proceedings of the 19th annual symposium on Integrated circuits and systems design, pages 38–43, New York, NY, USA. ACM.

Suen, T. and Wong, J. (1992). Efficient task migration algorithm for distributed systems. IEEE Transactions on Parallel and Distributed Systems, 3(4).

Zaki, M. J., Li, W., and Parthasarathy, S. (1996). Customized dynamic load balancing for a network of workstations. In HPDC ’96: Proceedings of the 5th IEEE International Symposium on High Performance Distributed Computing, page 282, Washington, DC, USA. IEEE Computer Society.