Abstract
Data compression in caches has been studied from the performance and energy consumption points of view. In this paper, we study the possible benefits of operating with compressed operands. Operating in the compressed domain (i.e., executing instructions with compressed operands) as far as possible can offer several advantages. First, since it is less reliant on decompression, it may be possible to employ a fancier compression/decompression strategy, without the associated performance/power penalties, that can offer higher compression rates to further boost cache locality. Second, the transfers between cache and datapath can use fewer bits (to transmit codes instead of data values) and provide dynamic energy savings in the corresponding bus. In this paper, we demonstrate that there are cases where operations can be executed in the compressed domain, which can lead to performance improvements and energy savings. In particular, we show that by operating in the compressed domain, the effectiveness of prior techniques can be further improved.
This research is partly supported by the NSF Career Award #0093082.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Abali, B., Franke, H.: Operating system support for fast hardware compression of main memory contents. In: Workshop on Solving the Memory Wall Problem (2000)
Benveniste, C., Franaszek, P., Robinson, J.: Cache-memory interfaces in compressed memory systems. In: Workshop on Solving the Memory Wall Problem (2000)
Bhandakar, D.: Alpha Implementations and Architecture. Complete Reference and Guide. Digital Press (1996)
Brooks, D., Martonosi, M.: Dynamically exploiting narrow width operands to improve processor power and performance. In: Proceedings of the Fifth Intl. Symposium on High- Performance Computer Architecture (1999)
Burger, D., Austin, T.M.: The SimpleScalar tool set, version 2.0. Computer Architecture News, 13–25 (June 1997)
Debray, S., Evans, W.: Profile-guided code compression. In: Proceeding of the Conference on Programming Language Design and Implementation, PLDI (2002)
Graefe, G., Shapiro, L.: Data compression and database performance. In: Proceedings of ACM/IEEE-CS Symposium On Applied Computing (1991)
Kirovsky, D., Kin, J., Mangione-Smith, W.H.: Procedure based program compression. In: Proceedings of the 30th International Symposium on Computer Architecture (1997)
Lee, J.-S., Hong, W.-K., Kim, S.-D.: Design and evaluation of a selective compressed memory system. In: Proceedings of the IEEE International Conference on Computer Design (1999)
Lefurgy, C., Bird, P., Chen, I.-C., Mudge, T.: Evaluation of a high performance code compression method. In: Proceedings of the 33th International Symposium on Computer Architecture (1997)
Lefurgy, C., Piccininni, E., Mudge, T.: Improving code density using compression techniques. In: Proceedings of the 32th International Symposium on Microarchitecture (1999)
Villa, L., Zhang, M., Asanović, K.: Dynamic zero compression for cache energy reduction. In: Proceedings of the 33rd International Symposium in Microarchitecture (2000)
Yang, J., Gupta, R.: Data compression transformations for dynamically allocated data structures. In: Proceedings of the International Conference on Compiler Construction (2002)
Yang, J., Gupta, R.: Energy efficient frequent value data cache design. In: Proceedings of the 35th Symposium on Microarchitecture (2002)
Yang, J., Zhang, Y., Gupta, R.: Frequent value compression in data caches. In: Proceedings of the 33rd Symposium on Microarchitecture (2000)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
De La Luz, V., Kandemir, M., Sivasubramaniam, A., Irwin, M.J. (2004). Exploring the Possibility of Operating in the Compressed Domain. In: Danelutto, M., Vanneschi, M., Laforenza, D. (eds) Euro-Par 2004 Parallel Processing. Euro-Par 2004. Lecture Notes in Computer Science, vol 3149. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-27866-5_67
Download citation
DOI: https://doi.org/10.1007/978-3-540-27866-5_67
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-22924-7
Online ISBN: 978-3-540-27866-5
eBook Packages: Springer Book Archive