Advertisement

An Energy Efficient Embedded Processor for Hard Real-Time Java Applications

  • Manish TewaryEmail author
  • Avinash Malik
  • Zoran Salcic
  • Morteza Biglari-Abhari
Conference paper
  • 519 Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11479)

Abstract

Energy management is very important and sometimes critical for certain classes of hard real-time systems. In this paper, we present effective energy reduction techniques for hard real-time systems developed in Java, which execute on bare metal and run on a time-predictable specialized Java processor. We modified traditional clock gating and dynamic frequency scaling methods to include the hardware-based run-time slack calculation in periodic tasks, thus reducing energy consumption in hard real-time systems. Two methods for energy reduction are employed leading to Energy Aware Java Optimized Processor (EAJOP). The first method includes task execution time monitoring and comparison with the estimated worst-case execution time to calculate the slack and bringing the processor to sleep for the slack duration upon task completion. The second method introduces real-time residual slack calculation at so-called checkpoints inside the periodic task, which are then used to lower the system frequency of the rest of the task dynamically, resulting in lower energy consumption. We compare EAJOP with baseline JOP when implemented on FPGA and demonstrate gains in energy consumption.

Keywords

Real-time and embedded systems Processor Compiler Energy management 

References

  1. 1.
    Schoeberl, M.: A Java processor architecture for embedded real-time systems. J. Syst. Archit. 54(1–2), 265–286 (2008)CrossRefGoogle Scholar
  2. 2.
    Schoeberl, M., Pedersen, R.: WCET analysis for a Java processor. In: Proceedings of JTRES 4th International Workshop on Java Technologies for Real Time and Embedded Systems, Paris (2006)Google Scholar
  3. 3.
    Huang, P.K., Ghiasi, S.: Efficient and scalable compiler directed energy optimizations for real time applications. ACM Trans. Des. Autom. Electron. Syst. 12(3), 27 (2007)CrossRefGoogle Scholar
  4. 4.
    Vijay, D., Tim, T.: Methodology for high level estimation of FPGA power consumption. In: Proceedings of the 2005 Asia and South Pacific Design Automation Conference, Shanghai (2005)Google Scholar
  5. 5.
    Sultan, H., Ananthanarayanan, G., Sarangi, S.R.: Processor power estimation techniques: a survey. Int. J. High Perform. Syst. Archit. 5(2), 93–114 (2014)CrossRefGoogle Scholar
  6. 6.
    Schoeberl, M., Preusser, T.B., Uhrig, S.: The embedded Java benchmark suite JemBench. In: JTRES’10 Proceedings of the 8th International Workshop on Java Technologies for Real-Time and Embedded Systems, Prague (2010)Google Scholar
  7. 7.
    Unsal, O.S., Koren, I.: System-level power-aware design techniques in real-time systems. In: Proceedings of the IEEE (2003)Google Scholar
  8. 8.
    Aboughazaleh, N., Mosse, D., Childers, B.R., Melhem, R.: Collaborative operating system and compiler power management for real-time applications. ACM Trans. Embed. Comput. Syst. 5(1), 82–115 (2006)CrossRefGoogle Scholar
  9. 9.
    Takase, H., Zeng, G., Gautheir, L., Kawashima, H.: An integrated optimization framework for reducing the energy consumption of embedded real-time applications. In: Proceedings of the 17th IEEE/ACM International Symposium on Low-Power Electronics and Design (2011)Google Scholar
  10. 10.
    Tatematsu, T., Takase, H., Gang, J., Tomiyama, H.: Checkpoint extraction using execution traces for intra-task DVFS in embedded systems. In: Sixth IEEE International Symposium on Electronic Design, Test and Application (2011)Google Scholar
  11. 11.
    Yuan, C., Schmitz, M.T., Al-hashimi, B.M., Reddy, S.M.: Workload-ahead-driven online energy minimization techniques for battery-powered embedded systems with time-constraints. ACM Trans. Des. Autom. Electron. Syst. (TODAES) 12(1), 19–24 (2007)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Manish Tewary
    • 1
    Email author
  • Avinash Malik
    • 1
  • Zoran Salcic
    • 1
  • Morteza Biglari-Abhari
    • 1
  1. 1.Department of Electrical and Computer EngineeringUniversity of AucklandAucklandNew Zealand

Personalised recommendations