AMNESIAC: Amnesic Automatic Computer
Pages 811 - 824
Abstract
Due to imbalances in technology scaling, the energy consumption of data storage and communication by far exceeds the energy consumption of actual data production, i.e., computation. As a consequence, recomputing data can become more energy efficient than storing and retrieving precomputed data. At the same time, recomputation can relax the pressure on the memory hierarchy and the communication bandwidth. This study hence assesses the energy efficiency prospects of trading computation for communication. We introduce an illustrative proof-of-concept design, identify practical limitations, and provide design guidelines.
References
[1]
ABRAHAM, S. G., SUGUMAR, R. A., WINDHEISER, D., RAU, B. R., AND GUPTA, R. Predictability of Load/Store Instruction Latencies. In International Symposium on Microarchitecture (MICRO) (1993).
[2]
BAILEY, D. H., BARSZCZ, E., BARTON, J. T., BROWNING, D. S., CARTER, R. L., DAGUM, L., FATOOHI, R. A., FRED-ERICKSON, P. O., LASINSKI, T. A., SCHREIBER, R. S., SIMON, H. D., VENKATAKRISHNAN, V., AND WEERATUNGA, S. K. The NAS Parallel Benchmarks: Summary and Preliminary Results. In Conference on Supercomputing (SC) (1991).
[3]
BERGMAN, K., BORKAR, S., CAMPBELL, D., CARLSON, W., DALLY, W., DENNEAU, M., FRANZON, P., HARROD, W., HILLER, J., AND KARP, S. Exascale Computing Study: Technology Challenges in Achieving Exascale Systems. DARPA Information Processing Techniques Of.ce (IPTO) sponsored study (2008).
[4]
BIENIA, C., KUMAR, S., SINGH, J. P., AND LI, K. The PARSEC Benchmark Suite: Characterization and Architectural Implications. Tech. Rep. TR-811-08, Princeton University, 2008.
[5]
BURGER, D., KAXIRAS, S., AND GOODMAN, J. R. Datascalar Architectures. In International Symposium on Computer Architecture (ISCA) (1997).
[6]
CARLSON, T. E., HEIRMAN, W., ALLAM, O., KAXIRAS, S., AND EECKHOUT,L. The Load Slice Core Microarchitecture. In International Symposium on Computer Architecture (ISCA) (2015).
[7]
CARLSON, T. E., HEIRMAN, W., AND EECKHOUT, L. Sniper: Exploring the Level of Abstraction for Scalable and Accurate Parallel Multi-core Simulation. In International Conference for High Performance Computing, Networking, Storage and Analysis (2011).
[8]
CHE, S., BOYER, M., MENG, J., TARJAN, D., SHEAFFER, J. W., LEE, S.-H., AND SKADRON, K. Rodinia: A Benchmark Suite for Heterogeneous Computing. In International Symposium onWorkload Characterization (2009).
[9]
COLLINS, J. D., WANG, H., TULLSEN, D. M., HUGHES, C., LEE, Y.-F., LAVERY, D., AND SHEN, J. P. Speculative Precomputation: Long-range Prefetching of Delinquent Loads. In International Symposium on Computer Architecture (ISCA) (2001).
[10]
DE KRUIJF, M., AND SANKARALINGAM, K. Idempotent Processor Architecture. In International Symposium on Microarchitecture (MICRO) (2011).
[11]
GONZALEZ, R., AND HOROWITZ, M. Energy Dissipation in General Purpose Microprocessors. IEEEJournal of Solid-State Circuits 31,9(1996).
[12]
GUO,X.,IPEK,E., AND SOYATA,T. Resistive Computation: Avoiding the Power Wall with Low-leakage, STT-MRAM Based Computing. In International Symposium on Computer Architecture (ISCA) (2010).
[13]
HENNING, J. L. SPEC CPU2006 Benchmark Descriptions. SIGARCH Computer Architecture News 34,4 (2006).
[14]
HOROWITZ, M. Computing's Energy Problem (and what we can do about it). Keynote at International Conference on Solid State Circuits (2014).
[15]
HU, Z., KAXIRAS, S., AND MARTONOSI, M. Timekeeping in the Memory System: Predicting and Optimizing Memory Behavior. In International Symposium on Computer Architecture (ISCA) (2002).
[16]
KANDEMIR, M., LI, F., CHEN, G., CHEN, G., AND OZTURK, O. Studying Storage-Recomputation Tradeoffs in Memory-Constrained Embedded Processing. In Design, Automation andTestin Europe(DATE) (2005).
[17]
KANG, Y., HUANG, W., YOO, S.-M., KEEN, D., GE, Z., LAM, V., PATTNAIK, P., AND TORRELLAS, J. FlexRAM: Toward an Advanced Intelligent Memory System. In International Conference on Computer Design (ICCD) (1999).
[18]
KECKLER, S. W., DALLY, W. J., KHAILANY, B., GARLAND, M., AND GLASCO, D. GPUs and the Future of Parallel Computing. IEEE Micro 31,5 (2011).
[19]
KOC, H.,KANDEMIR, M., ERCANLI, E., AND OZTURK, O. Reducing Off-Chip Memory Access Costs Using Data Recomputation in Embedded Chip Multi-processors. In Design Automation Conference (DAC) (2007).
[20]
KOC, H., OZTURK, O., KANDEMIR, M., AND ERCANLI, E. Minimizing Energy Consumption of Banked Memories Using Data Recomputation. In International Symposium on LowPower Electronics and Design (ISLPED) (2006).
[21]
KOGGE, P., BASS, S., BROCKMAN, J., CHEN, D., AND SHA,E. Pursuinga Peta.op: Point Designsfor100TF Computers Using PIM Technologies. In Frontiers of Massively Parallel Computing (1996).
[22]
KOGGE, P. M. The EXECUBE Approach to Massively Parallel Processing. In International Conference on Parallel Processing (ICPP) (1994).
[23]
LI, S., AHN, J. H., STRONG, R. D., BROCKMAN, J. B., TULLSEN,D.M., AND JOUPPI,N.P. McPAT:An Integrated Power, Area, andTiming Modeling Framework for Multicore and Manycore Architectures. In International Symposium on Microarchitecture (MICRO) (2009).
[24]
LIPASTI, M. H., WILKERSON, C. B., AND SHEN, J. P. Value Locality and Load Value Prediction. In International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) (1996).
[25]
LUK, C.-K., COHN, R., MUTH, R., PATIL, H., KLAUSER, A., LOWNEY, G., WALLACE, S., REDDI, V. J., AND HAZELWOOD, K. Pin: Building Customized Program Analysis Tools with Dynamic Instrumentation. In Conference on Programming Language Design and Implementation (PLDI) (2005).
[26]
MIGUEL,J.S.,BADR,M., AND JERGER,N.E. LoadValue Approximation. In International Symposium on Microarchitecture (MICRO) (2014).
[27]
MOSHOVOS, A., PNEVMATIKATOS, D. N., AND BANIASADI,A. Slice-processors:An Implementationof Operation-based Prediction. In International Conference on Supercomputing (ICS) (2001).
[28]
MOWRY, T. C., LAM, M. S., AND GUPTA, A. Design and Evaluation of a Compiler Algorithm for Prefetching. In International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) (1992).
[29]
OSKIN,M.,CHONG,F., AND SHERWOOD,T. ActivePages: a Computation Model for Intelligent Memory. In International Symposium on Computer Architecture (ISCA) (1998).
[30]
PATTERSON, D., ANDERSON, T., CARDWELL, N., FROMM, R., KEETON, K., KOZYRAKIS, C., THOMAS, R., AND YELICK, K. A Case for Intelligent RAM. IEEE Micro 17,2 (1997).
[31]
RIXNER, S., DALLY, W., KAPASI, U., KHAILANY, B., LOPEZ-LAGUNAS, A., MATTSON, P., AND OWENS, J. A Bandwidth-ef.cient Architecture for Media Processing. In International Symposium on Microarchitecture (MICRO) (1998).
[32]
ROTH, A., AND SOHI, G. S. A quantitative framework for automated pre-execution thread selection. In International Symposium on Microarchitecture (MICRO) (2002).
[33]
SHAO, Y., AND BROOKS, D. Energy Characterization and Instruction-Level Energy Model of Intel's Xeon Phi Processor. In International Symposium on Low Power Electronics and Design (ISLPED) (2013).
[34]
SODANI, A., AND SOHI, G. S. Dynamic Instruction Reuse. In International Symposium on Computer Architecture (ISCA) (1997).
[35]
STONE, H. S. A Logic-in-Memory Computer. IEEE Transactions on Computers C-19,1 (1970).
[36]
SUNDARAMOORTHY, K., PURSER, Z., AND ROTENBURG, E. Slipstream Processors: Improving Both Performance and FaultTolerance. InInternational Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) (2000).
[37]
ZILLES, C., AND SOHI, G. Execution-based Prediction Using Speculative Slices. In International Symposium on Computer Architecture (ISCA) (2001).
Recommendations
AMNESIAC: Amnesic Automatic Computer
ASPLOS '17Due to imbalances in technology scaling, the energy consumption of data storage and communication by far exceeds the energy consumption of actual data production, i.e., computation. As a consequence, recomputing data can become more energy efficient ...
AMNESIAC: Amnesic Automatic Computer
ASPLOS '17: Proceedings of the Twenty-Second International Conference on Architectural Support for Programming Languages and Operating SystemsDue to imbalances in technology scaling, the energy consumption of data storage and communication by far exceeds the energy consumption of actual data production, i.e., computation. As a consequence, recomputing data can become more energy efficient ...
E-MiLi: energy-minimizing idle listening in wireless networks
MobiCom '11: Proceedings of the 17th annual international conference on Mobile computing and networkingWiFi interface is known to be a primary energy consumer in mobile devices, and idle listening (IL) is the dominant source of energy consumption in WiFi. Most existing protocols, such as the 802.11 power-saving mode (PSM), attempt to reduce the time ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
April 2017856 pagesISBN:9781450344654DOI:10.1145/3037697- General Chairs:
- Yunji Chen,
- Olivier Temam,
- Program Chair:
- John Carter
Copyright © 2017 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 04 April 2017
Published in SIGARCH Volume 45, Issue 1
Check for updates
Author Tags
Qualifiers
- Research-article
Funding Sources
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations11Total Citations
- 645Total Downloads
- Downloads (Last 12 months)125
- Downloads (Last 6 weeks)15
Reflects downloads up to 06 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in