ABSTRACT
Talisman is a simulator that models the execution semantics and timing of a multicomputer. Talisman is unique in combining high semantic accuracy, high timing accuracy, portability, and good performance. This good performance allows users to run significant programs on large simulated multicomputers. The combination of high accuracy and good performance yields an ideal tool for evaluating architectural trade-offs. Talisman models the semantics of virtual memory, a circuit-switched internode interconnect, I/O devices, and instruction execution in both user and supervisor modes. It also models the timing of processor pipelines, caches, local memory buses, and a circuit-switched interconnect. Talisman executes the same program binary images as a hardware prototype at a cost of about 100 host instructions per simulated instruction. On a suite of accuracy benchmarks run on the hardware and the simulator, Talisman and the prototype differ in reported running times by only a few percent.
- 1.Robert C. Bedichek. Some efficient architecture simulation techniques. In Proceedings of the Winter 1990 USENIX Conference, pages 53-63, January 1990.Google Scholar
- 2.Robert C. Bedichek. The Meerkat Multicomputer: Trade-o.ffs in Mutticomputer Design. PhD thesis, University of Washington, August 1994. Department of Computer Science technical report 94-06-06 Google ScholarDigital Library
- 3.Jam~ R. lB~ll. Threaded code. Communicati~ms o/'the ACM (CACM), 16(2):370-372, June 1973. Google ScholarDigital Library
- 4.Eric A. Brewer, Chrysanthos N. Dellacrocas, Adrian Colbrook, and William E Weihl. PROTEUS: A hlgh-performanceparallel-architecture s~mulator. Technical Report MIT/LCS/TR-516, Massachusetts Institute of Technology, 1991. Google ScholarDigital Library
- 5.Enc A Brewer andWilliam E Weihl Developmg parallel applications using high-performance simulation. ACM/ONR Workshop on Parallel and Dist Debugging ACM SIGPLAN Notices, 28(12):158-168, December 1993. Google ScholarDigital Library
- 6.J. Bradley Chen and Bnan N Bershad. The impact of operating system performance on memory system performance Proceedings o{ the 14th ACM Sympostum on Operating System Principles, pages 120-133, December 1993 Google ScholarDigital Library
- 7.D.W. Clark. Pipelinlng and performance in the VAX-8800 processor. Symposium on A rc htte c tu raI Suppo rt f o r P ro g rammm g Languages and Operating Systems, October 1987. Google ScholarCross Ref
- 8.Robert E Cmelik and David Keppel. Shade: A fast instruction-set simulator for execution profiling. In Proceedings of the 1994 ACM SIGMETRICS Conference on Modeling and Measurement of Computer Systems, May 1994. Google ScholarDigital Library
- 9.R C Covington, S Madala, V Mehta, J.R. Jump, and J.B. Sinclair. The Rice parallel processing testbed. In Proceedings of the 1988 ACM SIGMETRICS and PERFORMANCE Conference, pages 4-11, May 1988. Google ScholarDigital Library
- 10.Peter Deutsch and Alan M Schiffman. Efficient implementation of the Smalltalk-80 system l l th Annual Symposium on PrinctpIes of Programming Languages, pages 297-302, January 1984. Google ScholarDigital Library
- 11.David R. Ditzel, Hubert R. MeLellan, and Alan D. Berenbaum The hardware architecture of the CRISP microprocessor In Proceedings of the 14th Annual International Symposium on Computer Architecture; Computer Architecture News, pages 309-319, June 1987 Google ScholarDigital Library
- 12.R. M. Fujimoto and W. B. Campbell. Efficient Instruction Level Simulation of Computers. Transacttons of the Soctety for Computer Stmulatton, 5(2):109-124, Apr. 1988. Google ScholarDigital Library
- 13.Stephen R. Goldschmidt. Stmualtton of Multtprocessors: Accuracy and Pertormance. PhD thesis, Stanford University, June 1993. Google ScholarDigital Library
- 14.TG Lang, J.T. O'Quin, and R O Simpson. Threaded code interpreter for object code IBM Technical Disclosure Bulletin, pages 4238-4241, March 1986Google Scholar
- 15.Peter Magnusson Efficient simulation of parallel hardware Masters thesis. Royal Institute of Technology (KTH), Stockholm, Sweden, 1992Google Scholar
- 16.Peter Magnusson and Bengt Werner. Efficient memory simulation in StmlCS In Proceedings of the 28th Annual Stmulatton Sympostum, 1995. Google ScholarDigital Library
- 17.Peter S. Magnusson. A design for efficient simulation of a mult~processor MASCOTS '93 - Proceedmgs oJ the 1993 Western Stmutatton Multtconference on International Workshop on Modeling, Analysts, and StmuIatton of Computer and Telecommuntcatton Systems, January 1993. Google ScholarDigital Library
- 18.Henry Massahn and Calton Pu. Threads and input/output m the Synthesis kernel. In Proceedings of the twelth ACM Sympostum on operating system principles, pages 191-201, December 1989. Google ScholarDigital Library
- 19.Cathy May Mimic A fast S/370 simulator. In Proceedings o} the ACM SIGPLAN 1987 Symposium on Interpreters and Interpretive Techniques, SIGPLAN Notices, volume 22, pages 1-13, St. Paul, Minnesota, June 1987 Google ScholarDigital Library
- 20.MC88100 RISC Mtcroprocessor User's Manual. Motorola Corporation, 2900 South Dmblo Way, Tempe, Arizona,Google Scholar
- 21.MC88200 Cache~Memory Management User's Manual. Motorola Corporation, 2900 South Diablo Way, Tempe, Arizona.Google Scholar
- 22.Carl Ponder Personal communication, February 1994Google Scholar
- 23.Steven K. Reinhardt, Mark D, Hill, James R. Larus, Alvin R. Lebeck, James C. Lewis, and David A. Wood. The Wisconsin Wind Tunnel Virtual prototyping of parallel computers. Per~ormance Evaluatton Review, 21(l)'48-60, May 1993. Google ScholarDigital Library
- 24.Mendel Rosenblum and Emmett Witchel. SimOS. A Platform for Complete Workload Studies. Personal Communication (to be published), 1995Google Scholar
- 25.Margo Selzer, Peter Chen, and John Ousterhout. D~sk scheduling revisited. In Proceedings of the Wmter 1990 USENIX Conference, pages 313-324, January 1990.Google Scholar
- 26.Richard L. Sites, Anton Chernoff, Mathew B. Kerk, Maurice P Marks, and Scott G. Robinson Binary translation. Commumcattons of the ACM, pages 69-81, February 1993. Google ScholarDigital Library
- 27.Insignia Solutions. SoflPC Product Information, 1991.Google Scholar
- 28.Richard M. Stallman and Roland H. Pesch Using GDB: The GNU Source-LevelDebugger. Free Software Foundation, 545 Tech Square, Cambridge, Ma. 02139, March 1992Google Scholar
Index Terms
- Talisman: fast and accurate multicomputer simulation
Recommendations
Talisman: fast and accurate multicomputer simulation
Talisman is a simulator that models the execution semantics and timing of a multicomputer. Talisman is unique in combining high semantic accuracy, high timing accuracy, portability, and good performance. This good performance allows users to run ...
Architecture and design of a Talisman-compatible multimedia processor
This paper describes the architecture, functionality, and design of a Talisman-compatible multimedia processor (TM-PC) from Philips Semiconductors. “Talisman” is the code name of a new graphics and multimedia hardware architecture (from ...
A High-Throughput Distributed Shared-Buffer NoC Router
Microarchitectural configurations of buffers in routers have a significant impact on the overall performance of an on-chip network (NoC). This buffering can be at the inputs or the outputs of a router, corresponding to an input-buffered router (IBR) or ...
Comments