The method for providing quality of service time requirements in reconfigurable computing systems

Authors

DOI:

https://doi.org/10.15587/1729-4061.2016.81003

Keywords:

reconfigurable computing systems, quality of service, reconfiguration overheads, Field-Programmable Gate Array (FPGAs)

Abstract

The process of calculations control in reconfigurable computing systems characterized by the time constraints of both the applications solved and the functional and structural organization of the system is investigated.

As part of this, the method is modified and mathematically substantiated for determining the time intervals of tasks. The proposed modification to the analysis of tasks acceleration index allows to evaluate the unproductive time expenditure taking into account the FPGA hardware limitations in the computer algorithms mapping to the reconfigurable computing structure.

The method of providing time parameters of service quality in the reconfigurable computing system that provides given application runtime by determining the amount of unproductive time-consuming reconfiguration and the choice of optimal service discipline for each task in terms of time and existing hardware limitations was proposed. The use of the proposed method of determining the time intervals, which the implementation of the method is based on determining the sequence of tasks for which the target architecture of the computer system is effective and the use of additional mechanisms to reduce overhead does not speed up calculations within the advertised time requirements. Application of the proposed method allows to determine and optimize the impact of space limitations the reconfiguration time and reduce the number of rejections of tasks in the dynamic mapping of task flows to the reconfigurable computational structure. The formalization of the method for determining the time intervals of tasks was developed, for which the optimal boundaries of the effective use of the proposed facilities in terms of the adequacy of the offered application time limits to the limits of the reconfigurable computer system were defined and justified.

The proposed tools can be used in modern high-performance reconfigurable computing systems during solving the problems of control of various technical and management processes and implementation of multidimensional calculations in complex information systems.

Author Biographies

Iryna Klymenko, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” Peremohy ave., 37, Kyiv, Ukraine, 03056

PhD, Associate Professor

Department of Computer Engineering

Yurii Kulakov, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” Peremohy ave., 37, Kyiv, Ukraine, 03056

Doctor of Technical Sciences, Professor

Department of Computer Engineering

Valentyna Tkachenko, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” Peremohy ave., 37, Kyiv, Ukraine, 03056

PhD, Associate Professor

Department of Computer Engineering

Oleksandr Storozhuk, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” Peremohy ave., 37, Kyiv, Ukraine, 03056

Postgraduate student

Department of Computer Engineering

References

  1. Ahmed, W., Shafique, M., Bauer, L., Henkel, J. (2011). Adaptive resource management for simultaneous multitasking in mixed-grained reconfigurable multi-core processors. Proceedings of the Seventh IEEE/ACM/IFIP International Conference on Hardware/software Codesign and System Synthesis - CODES+ISSS ’11, 365–374. doi: 10.1145/2039370.2039426
  2. Huang, M., Narayana, V. K., Simmler, H., Serres, O., El-Ghazawi, T. (2010). Reconfiguration and Communication-Aware Task Scheduling for High-Performance Reconfigurable Computing. ACM Transactions on Reconfigurable Technology and Systems, 3 (4), 1–25. doi: 10.1145/1862648.1862650
  3. Bassiri, M. M., Shahriar, S. H. (2010). Mitigating reconfiguration overhead in on-line task scheduling for reconfigurable computing systems. 2010 2nd International Conference on Computer Engineering and Technology, 397–402. doi: 10.1109/iccet.2010.5485509
  4. Al-Wattar, A., Areibi, S., Saffih, F. (2012). Efficient On-line Hardware/Software Task Scheduling for Dynamic Run-time Reconfigurable Systems. 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum, 401–406. doi: 10.1109/ipdpsw.2012.50
  5. Taher, M., El-Ghazawi, T. (2009). Virtual Configuration Management: A Technique for Partial Runtime Reconfiguration. IEEE Transactions on Computers, 58 (10), 1398–1410. doi: 10.1109/tc.2009.81
  6. Liu, S., Pittman, R. N., Forin, A., Gaudiot, J.-L. (2013). Achieving energy efficiency through runtime partial reconfiguration on reconfigurable systems. ACM Transactions on Embedded Computing Systems, 12 (3), 1–21. doi: 10.1145/2442116.2442122
  7. Kulakov, Y. O., Klymenko, I. A. (2015). The multilevel memory organisation in the reconfigurable computing systems. Visnyk NTUU “KPI”. Informatyka, upravlinnia ta obtchislyuvalna technika, 61, 18–26.
  8. Ahmed, W., Shafique, M., Bauer, L., Henkel, J. (2011). mRTS: Run-time system for reconfigurable processors with multi-grained instruction-set extensions. 2011 Design, Automation & Test in Europe, 1–6. doi: 10.1109/date.2011.5763246
  9. Koenig, R., Bauer, L., Stripf, T., Shafique, M., Ahmed, W., Becker, J., Henkel, J. (2010). KAHRISMA: A Novel Hypermorphic Reconfigurable-Instruction-Set Multi-grained-Array Architecture. 2010 Design, Automation & Test in Europe Conference & Exhibition (DATE 2010), 819–824. doi: 10.1109/date.2010.5456939
  10. Dümmler, J., Rauber, T., Rünger, G. (2009). Scalable computing with parallel tasks. Proceedings of the 2nd Workshop on Many-Task Computing on Grids and Supercomputers - MTAGS’09, 1–10. doi: 10.1145/1646468.1646477
  11. Klymenko, I. A. (2015). The effectiveness analysis of resources management in reconfigurable computer systems. Visnyk NTUU “KPI”. Informatyka, upravlinnia ta obtchislyuvalna technika, 62, 11–21.
  12. Kulakov, Y. O., Klymenko, I. A., Rudnytskyi, M. V. (2015). Development of the reconfiguration acceleration method in the dynamically reconfigurable computing systems. Eastern-European Journal of Enterprise Technologies, 4 (4 (76)), 25–29. doi: 10.15587/1729-4061.2015.47227

Downloads

Published

2016-10-30

How to Cite

Klymenko, I., Kulakov, Y., Tkachenko, V., & Storozhuk, O. (2016). The method for providing quality of service time requirements in reconfigurable computing systems. Eastern-European Journal of Enterprise Technologies, 5(9 (83), 4–12. https://doi.org/10.15587/1729-4061.2016.81003

Issue

Section

Information and controlling system