On the Effectiveness of the Scheduling Algorithm of the Dynamically Trace Scheduled VLIW
Resumo
In a machine that follows the dynamically trace scheduled VLIW (DTSVLIW) architecture, VLIW instructions are built dynamically through a scheduling algorithm that can be implemented in hardware. These VLIW instructions are cached so that the machine can spend most of its time executing VLIW instructions without sacrificing any binary compatibility. This paper evaluates the effectiveness of the DTSVLIW instruction scheduling algorithm by comparing it with the first come first served (FCFS) algorithm, used for microinstruction compaction, and the Greedy algorithm, used by the Dynamic Instruction Formatting architecture. In order to perform these comparisons, we have performed experiments using the SPECint95 benchmark suite.
Referências
A. F. de Souza and P. Rounce, "Dynamically Trace Scheduled VLIW Architectures". Proceedings of The HPCN'98, in Lecture Notes on Computer Science, Vol. 1401, pp. 993-995, 1998.
A. F. de Souza and P. Rounce, "SPECint95 Performance of an Implementation of the Dynamically Trace Scheduled VLIW Architecture", Proceedings of The 10th Brazilian Symposium on Computer Architecture and High Performance Computing, pp. 185-188, 1998.
A. F. de Souza and P. Rounce, "Dynamically Scheduling the Trace Produced During Program Execution into VLIW Instructions", Proceedings of The 2nd IPPS/SPDP Merged Symposium: 13th International Parallel Processing Symposium & 10th Symposium on Parallel and Distributed Processing, pp. 248-257, 1999.
A. F. de Souza and P. Rounce, "Effect of Multicycle Instructions on the Integer Performance of the Dynamically Trace Scheduled VLIW Architecture", Proceedings of The HPCN'99, in Lecture Notes on Computer Science, Vol. 1593, pp. 1203-1206. 1999.
D. J. DeWitt, "A Machine Independent Approach to the Production of Optimal Horizontal Microcode", Technical Report 76 DT4, University of Michigan, Ann Arbor, August 1976.
K. Ebcioglu and E. R. Altman, "DAISY: Dynamic Compilation for 100% Architectural Compatibility", Proceedings of The 24th International Symposium on Computer Architecture, pp. 26-37. 1997.
J. A. Fisher, 'The VLIW Machine: A Multiprocessor for Compiling Scientific Code", IEEE Computer, pp. 45-53, July I 984.
T. Hara, et al., "Performance Comparison of ILP Machines with Cycle Time Evatuation", Proceedings of The 23rd International Symposium on Computer Architecture, pp. 18-26, 1996.
W. W. Hwu, and Y. N. Pau, "Checkpoint Repair for Out-of-order Execution Machines", Proceedings of The 14th International Symposium on Computer Architecture, pp. 18-26. 1987.
R. Nair, M. E. Hopkins, "Exploiting Instructions Levei Parallelism in Processors by Caching Scheduled Groups", Proceedings of The 24th International Symposium on Computer Architecture, pp. 13-25, 1997.
B. R. Rau. "Dynamically Scheduled VLIW Processors", Proceedings of The 26th International Symposium on Microarchitccture, pp. 80-92, 1993.
E. Rotenberg, et al. "Trace Processors". Proceedings of The 30th International Symposium on Microarchitecture, pp. 138-148, 1997.
J. E. Smith and S. Weiss, "PowerPC 601 and A1pha 21064: A Tale of Two RISCs," IEEE Computer, pp. 46-58, June 1994.
A. Sodani and G. Sohi, "Understanding the Difference Between Value Prediction and Instruction Reuse", Proceedings of The 31st International Symposium on Microarchitecture, pp. 205-215, 1998.
Sun Microsystems, 'The Sparc Architecture Manual - Version 7", Sun Microsystems Inc., 1987.