Trace Cache performance parameters

Instruction fetch mechanism is a performance bottleneck of a Superscalar Processor. The fetch performance of the processor can be improved with the aid of an instruction memory structure known as Trace Cache. This paper presents parameters and analytical expressions, which describe instruction fetch performance of a Trace Cache microarchitecture. The instruction fetch rates predicted by the expressions differ by seven percent from the simulated fetch rates for SPEC2000 benchmark programs. Presented analytical expressions are implemented in a computer program named Tulip. Tulip is used to explore parameters, and their influence on fetch performance. Tulip is also used to understand Trace Cache performance tradeoffs.

[1]  Yale N. Patt,et al.  Trace cache design for wide-issue superscalar processors , 1999 .

[2]  Eric Rotenberg,et al.  A Trace Cache Microarchitecture and Evaluation , 1999, IEEE Trans. Computers.

[3]  Daniel J. Pease,et al.  An analytical model for trace cache instruction fetch performance , 2001, Proceedings 2001 IEEE International Conference on Computer Design: VLSI in Computers and Processors. ICCD 2001.

[4]  James E. Smith,et al.  Path-based next trace prediction , 1997, Proceedings of 30th Annual International Symposium on Microarchitecture.

[5]  Nader Bagherzadeh,et al.  Modeled and Measured Instruction Fetching Performance for Superscalar Microprocessors , 1998, IEEE Trans. Parallel Distributed Syst..

[6]  Jack L. Lo,et al.  Exploiting Choice: Instruction Fetch and Issue on an Implementable Simultaneous Multithreading Processor , 1996, 23rd Annual International Symposium on Computer Architecture (ISCA'96).

[7]  Dean M. Tullsen,et al.  Exploiting Choice: Instruction Fetch and Issue on an Implementable Simultaneous Multithreading Processor , 1996, 23rd Annual International Symposium on Computer Architecture (ISCA'96).