Energy reduction of the fetch mechanism through dynamic adaptation
Energy reduction of the fetch mechanism through dynamic adaptation
- Author(s): S. del Pino ; D. Chaver ; L. Pinuel ; M. Prieto ; F. Tirado
- DOI: 10.1049/iet-cdt:20060179
For access to this article, please select a purchase option:
Buy article PDF
Buy Knowledge Pack
IET members benefit from discounts to all IET publications and free access to E&T Magazine. If you are an IET member, log in to your account and the discounts will automatically be applied.
Thank you
Your recommendation has been sent to your librarian.
- Author(s): S. del Pino 1 ; D. Chaver 1 ; L. Pinuel 1 ; M. Prieto 1 ; F. Tirado 1
-
-
View affiliations
-
Affiliations:
1: ArTeCS Group, Complutense University of Madrid, Madrid, Spain
-
Affiliations:
1: ArTeCS Group, Complutense University of Madrid, Madrid, Spain
- Source:
Volume 2, Issue 2,
March 2008,
p.
94 – 107
DOI: 10.1049/iet-cdt:20060179 , Print ISSN 1751-8601, Online ISSN 1751-861X
A highly efficient fetch unit is essential not only to obtain good performance but also to achieve energy efficiency. However, existing commercial fetch designs are not adaptable and depending on the program behaviour, they can be either insufficient or an overkill. A phase-based adaptive fetch mechanism that can be dynamically adjusted based on feedback information of the program behaviour is introduced. This design adds very little hardware complexity and relegates complex tasks to the software components. It is also very effective: saving 35% and 52% fetch energy on an average compared with a conventional and a trace cache-based fetch unit, respectively. At the same time, performance is improved by 4.7% and 0.8%, respectively.
Inspec keywords: program compilers; cache storage
Other keywords:
Subjects: File organisation; Compilers, interpreters and other processors
References
-
-
1)
- Rotenberg, E., Bennett, S., Smith, J.E.: `Trace cache: a low latency approach to high bandwidth instruction fetching', Int. Symp. Microarchitecture, Novermber 1996.
-
2)
- T. Cormen , C. Leiserson , R. Rivest . (1989) Introduction to algorithms.
-
3)
- T. Austin , E. Larson , D. Ernst . SimpleScalar: an infrastructure for computer system modeling. Computer , 2
-
4)
- Powell, M., Yang, S.-H., Falsafi, B.: `Gated-Vdd: A circuit technique to reduce leakage in deep-submicron cache memories', Proc. 2000 Int. Symp. Low Power Electronics and Design, July 2000, p. 90–95.
-
5)
- Huang, M.C., Renau, J., Torrellas, J.: `Positional adaptation of processors: application to energy reduction', Int. Symp. Computer Architecture, June 2003.
-
6)
- Yang, S., Powell, M.D., Falsafi, B.: `An integrated circuit/architecture approach to reducing leakage in deep-submicron high-performance I-caches', Int. Symp. High-Performance Computer Architecture, January 2001.
-
7)
- Baniasadi, A., Moshovos, A.: `Instruction flow-based front-end throttling for power-aware high-performance processors', Int. Symp. Low Power Electronics and Design, August 2001, p. 16–21.
-
8)
- Zhang, Y., Parikh, D., Sankaranarayanan, K.: `HotLeakage: a temperature-aware model of subthreshold and gate leakage for architects', Technical Report CS-2003–05, March 2003.
-
9)
- Dhodapkar, A.S., Smith, J.E.: `Managing multiconfiguration hardware via dynamic working set analysis', Int. Symp. Computer Architecture, 2002, p. 233–244.
-
10)
- Hu, J.S., Vijaykrishnan, N., Irwin, M.J.: `Using dynamic branch behavior for power-efficient instruction fetch', Int. Symp. VLSI, February 2003.
-
11)
- Balasubramonian, R., Albonesi, D., Buyuktosunoglu, A.: `Memory hierarchy reconfiguration for energy and performance in general-purpose processor architectures', Int. Symp. Microarchitecture, December 1999, p. 245–257.
-
12)
- M. Co , K. Skadron . Evaluating trace caches energy-efficiency. ACM Trans. Archit. Code Optim. , 450 - 476
-
13)
- Manne, S., Klauser, A., Grunwald, D.: `Pipeline gating: speculation control for energy reduction', Int. Symp. Computer Architecture, July 1998, p. 132–141.
-
14)
- Brooks, D., Tiwari, V., Martonosi, M.: `Wattch: a framework for architectural-level power analysis and optimizations', Int. Symp. Computer Architecture, July 2001.
-
15)
- A. Buyuktosunoglu , T. Karkhanis , D.H. Albonesi . Energy efficient co-adaptive instruction fetch and issue. Comput. Archit. News
-
16)
- Liu, W., Huang, M.C.: `EXPERT: expedited simulation exploiting program behavior repetition', Int. Conf. Supercomputing, June 2004.
-
17)
- Chaver, D., Rojas, M.A., Pinuel, L.: `Energy-aware fetch mechanism: trace cache and BTB customization', Int. Symp. Low Power Electronics and Design, August 2005.
-
18)
- R. Bahar , S. Manne . Power and energy reduction via pipeline balancing. Int. Symp. Computer Architecture , 218 - 229
-
19)
- Karkhanis, T., Smith, J.E., Bose, P.: `Saving energy with just in time instruction delivery', Proc. Int. Symp. Low Power Electronics and Design, August 2002.
-
20)
- O.S. Unsal , R. Ashok , I. Koren . Cool-cache: a compiler-enabled energy efficient data caching framework for embedded/multimedia processors. ACM Trans. Embedded Comput. Syst. , 3 , 373 - 392
-
21)
- Unsal, O.S., Koren, I., Krishna, C.M.: `Cool-fetch: a compiler-enabled IPC estimation based framework for energy reduction', Workshop on Interaction between Compilers and Computer Architectures, February 2004, p. 43–52.
-
22)
- Yeh, T.Y., Patt, Y.N.: `Alternative implementations of two-level adaptive branch prediction', Int. Symp. Computer Architecture, May 1992.
-
23)
- Hu, J.S., Vijaykrishnan, N., Irwin, M.J.: `Optimizing power efficiency in trace cache fetch unit', Technical Report, 2003.
-
24)
- Makino, H., Tujihashi, Y., Morishima, C.: `A low power SRAM using auto-backgate-controlled MT-CMOS', Proc. 1998 Int. Symp. Low Power Electronics and Design, August 1998, p. 293–98.
-
25)
- H. Hanson , M.S. Hrishikesh , V. Agarwal . Static energy reduction techniques for microprocessor caches. IEEE Trans. VLSI Syst. , 3
-
26)
- M.C. Huang , D. Chaver , L. Pinuel . Customizing the branch predictor to reduce complexity and energy consumption. IEEE Micro , 5 , 12 - 25
-
27)
- D.H. Albonesi . Selective cache ways: on-demand cache resource allocation. J. Instr. Level Parallelism
-
1)