DVS for buffer-constrained architectures with predictable QoS-energy tradeoffs

ABSTRACT

We present a new scheme for dynamic voltage and frequency scaling (DVS) for processing multimedia streams on architectures with restricted buffer sizes. The main advantage of our scheme over previously published DVS schemes is its ability to provide hard QoS guarantees while still achieving considerable energy savings. Our scheme can handle workloads characterized by both, the data-dependent variability in the execution time of multimedia tasks and the burstiness in the on-chip traffic arising out of multimedia processing. Many previous DVS algorithms capable of handling such workloads rely on control-theoretic feedback mechanisms or prediction schemes based on probabilistic techniques. Usually it is difficult to provide QoS guarantees with such schemes. In contrast, our scheme relies on worst-case interval-based characterization of the workload. The main novelty of our scheme is a combination of offline analysis and runtime monitoring to obtain worst case bounds on the workload and then improving these bounds at runtime. Our scheme is fully scalable and has a bounded application-independent runtime overhead.

REFERENCES

• 1: Todd Austin , Eric Larson , Dan Ernst, SimpleScalar: An Infrastructure for Computer System Modeling, Computer, v.35 n.2, p.59-67, February 2002 [doi>10.1109/2.982917]
• 2: Jean-Yves Le Boudec , Patrick Thiran, Network calculus: a theory of deterministic queuing systems for the internet, Springer-Verlag New York, Inc., New York, NY, 2001
• 3: Vadim Gutnik , Anantha P. Chandrakasan, Embedded power supply for low-power DSP, IEEE Transactions on Very Large Scale Integration (VLSI) Systems, v.5 n.4, p.425-435, Dec. 1997 [doi>10.1109/92.645069]
• 4: Chaeseok Im , Soonhoi Ha , Huiseok Kim, Dynamic voltage scheduling with buffers in low-power multimedia applications, ACM Transactions on Embedded Computing Systems (TECS), v.3 n.4, p.686-705, November 2004 [doi>10.1145/1027794.1027796]
• 5: Y.-H. Lu, L. Benini, and G. De Micheli. Dynamic frequency scaling with buffer insertation for mixed workloads. IEEE Trans. on CAD of Integrated Circuits and Systems, 21(11), November 2002.
• 6: Zhijian Lu , John Lach , Mircea Stan , Kevin Skadron, Reducing Multimedia Decode Power using Feedback Control, Proceedings of the 21st International Conference on Computer Design, p.489, October 13-15, 2003
• 7: Tajana Simunic , Stephen P. Boyd , Peter Glynn, Managing power consumption in networks on chips, IEEE Transactions on Very Large Scale Integration (VLSI) Systems, v.12 n.1, p.96-107, January 2004 [doi>10.1109/TVLSI.2003.820533]
• 8: Qiang Wu , Philo Juang , Margaret Martonosi , Douglas W. Clark, Formal online methods for voltage/frequency control in multiple clock domain microprocessors, Proceedings of the 11th international conference on Architectural support for programming languages and operating systems, October 07-13, 2004, Boston, MA, USA

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.3 SPECIAL-PURPOSE AND APPLICATION-BASED SYSTEMS
      Subjects: Real-time and embedded systems

General Terms:
Algorithms, Design, Performance

Keywords:
DVS, QoS, buffer management, predictable design