Wisconsin Wind Tunnel II: a fast, portable parallel architecture simulator
IEEE ConcurrencyPublished 1 October 2000Open access
S. Mukherjee, S.K. Reinhardt, Babak Falsafi, Michael Litzkow, Mark D. Hill, David A. Wood
Citations165
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Through four key operations, the Wisconsin Wind Tunnel II can easily run simulations on Sparc platforms ranging from a workstation cluster to an asymmetric multiprocessor.
Abstract
To analyze new, parallel computers, developers must rapidly simulate designs running realistic workloads. Historically, direct execution and a parallel host have accelerated simulations, although these techniques have typically lacked portability. Through four key operations, the Wisconsin Wind Tunnel II can easily run simulations on Sparc platforms ranging from a workstation cluster to a symmetric multiprocessor.
Keywords
Computer ScienceDecision Sciences
ACM SIGARCH Computer Architecture NewsThe SimpleScalar tool set, version 2.0
3,064 Citations1997Doug Burger, Todd Austin
This document describes release 2.0 of the SimpleScalar tool set, a suite of free, publicly available simulation tools that offer both detailed and high-performance simulation of modern microprocessors.
Communications of the ACMParallel discrete event simulation
1,816 Citations1990Richard M. Fujimoto
This article deals with the execution of a simulation program on a parallel computer by decomposing the simulation application into a set of concurrently executing processes and introduces interesting synchronization problems that are at the heart of the PDES problem.
IEEE MicroMyrinet: a gigabit-per-second local area network
1,702 Citations1995N.J. Boden, D. Cohen +5 more
The Myrinet local area network employs the same technology used for packet communication and switching within massively parallel processors, but with the highest performance per unit cost of any current LAN.
Active messages
1,397 Citations1992Thorsten von Eicken, David Culler +2 more
It is shown that active messages are sufficient to implement the dynamically scheduled languages for which message driven machines were designed and, with this mechanism, latency tolerance becomes a programming/compiling concern.
EEL
415 Citations1995James R. Larus, Eric Schnarr
EEL supports a machine- and system-independent editing model that enables tool builders to modify an executable without being aware of the details of the underlying architecture or operating system or being concerned with the consequences of deleting instructions or adding foreign code.
The Wisconsin Wind Tunnel
353 Citations1993Steven K. Reinhardt, Mark D. Hill +4 more
A new technique for evaluating cache coherent, shared-memory computers and the Wisconsin Wind Tunnel (WWT) is developed, which correctly interleaves target machine events and calculates target program execution time.
ACM Transactions on Modeling and Computer SimulationUsing the SimOS machine simulator to study complex computer systems
347 Citations1997Mendel Rosenblum, Edouard Bugnion +2 more
This paper identifies two challenges that machine simulators such as SimOS must overcome in order to effectively analyze large complex workloads: handling long workload execution times and collecting data effectively.
PROTEUS: a high-performance parallel-architecture simulator
222 Citations1992Kyle Palos, Chrysanthos Dellarocas +2 more
PROTEUS is a high-performance simulator for MIMD multiprocessors that provides repeatability, nonintrusive monitoring and debugging, and integrated graphical output, which result in a development environment superior to those available on real multipROcessors.
International journal in computer simulationSPEEDES - A multiple-synchronization environment for parallel discrete-event simulation
149 Citations1992Jeff S. Steinman
The rules for processing events in SPEEDes are discussed, the implementation of various other synchronization protocols supported by SPEEDES is described, some new ones for the future are described, discusses interactive simulations, and some performance results are given.
SimICS/sun4m: a virtual workstation
126 Citations1998Peter Magnusson, Fredrik Dahlgren +8 more
The performance of SimICS/sun4m is sufficient to run realistic workloads, such as the database benchmark TPC-D, scaling factor 1/100, or an interactive network application such as Mozilla, and the slowdown in relation to native hardware is in the range of 25 to 75 (measured using SPECint95).
Reactive NUMA
109 Citations1997Babak Falsafi, David A. Wood
This paper proposes and evaluates a new approach to directory-based cache coherence protocols called Reactive NUMA (R-NUMA), which combines a conventional CC-N UMA coherence protocol with a more-recent Simple-COMA (S- COMA) protocol.
ACM SIGMETRICS Performance Evaluation ReviewThe rice parallel processing testbed
105 Citations1988R. C. Covington, S. Madala +3 more
Rice Digital Scholarship Archive (Rice University)RSIM: An Execution-Driven Simulator for ILP-Based Shared-Memory Multiprocessors and Uniprocessors
93 Citations1997Vijay S. Pai, Parthasarathy Ranganathan +1 more
RSIM is execution-driven and models state-of-the-art ILP processors, an aggressive memory system, and a multiprocessor coherence protocol and interconnect, including contention at all resources, as well as being used successfully in both undergraduate and graduate computer architecture courses at Rice University.
ComputerCost-effective parallel computing
88 Citations1995David A. Wood, M.D. Hill
How large memories can make parallel computing cost-effective even with modest speedups is considered, and actual costups can be far less than linear for systems with hundreds of Mbytes of main memory.
Proceedings of the International Conference on Parallel ProcessingMultiprocessor Simulation and Tracing Using Tango.
87 Citations1991Helen Davis, Stephen R. Goldschmidt +1 more
A novel method to locate hydrogen atoms in amino acids is demonstrated, which involves measuring the C(alpha)H( alpha) bond vector geometry through orientationally dependent dipolar coupling frequencies measured by Lee-Goldburg cross polarization (LGCP).
Coherent network interfaces for fine-grain communication
85 Citations1996Shubhendu S. Mukherjee, Babak Falsafi +2 more
This paper begins an exploration of network interfaces (NIs) that use coherence---coherent network interface (CNIs)---to improve communication performance, restricting this study to NI/CNIs that reside on coherent memory or I/O buses, to NI's that are much simpler than processors, and to the performance of fine-grain messaging from user process to user process.
Decoupled hardware support for distributed shared memory
83 Citations1996Steven K. Reinhardt, Robert W. Pfile +1 more
Hardware support for fine-grain distributed shared memory in networks of workstations is investigated to reduce design time and implementation cost relative to dedicated DSM systems, and two decoupled systems are presented, allowing greater use of off-the-shelf devices.
The impact of instruction-level parallelism on multiprocessor performance and simulation methodology
72 Citations2002Vijay S. Pai, Parthasarathy Ranganathan +1 more
This paper presents the first detailed analysis of the impact of such processors on shared-memory multiprocessors using a detailed execution-driven simulator, and examines the validity of common direct-execution simulation techniques that employ previous-generation processor models to approximate ILP-based multiprocessionors.
Minds at UW (University of Wisconsin)The Wisconsin Wind Tunnel: virtual prototyping of parallel computers
60 Citations1995Steven K. Reinhardt, Mark D. Hill +4 more
Tempest: a substrate for portable parallel programs
42 Citations2002M.D. Hill, James R. Larus +1 more
The paper describes the Tempest mechanisms, briefly explains how they are used, outlines several implementations on both custom and stock hardware, and presents preliminary performance results that demonstrate the benefits of this approach.
ACM Transactions on Modeling and Computer SimulationModeling cost/performance of a parallel computer simulator
40 Citations1997Babak Falsafi, David A. Wood
A performance model of the Wisconsin Wind Tunnel, a system that simulates cache-coherent shared-memory machines on a message-passing Thinking Machines CM-5, indicates that for target system sizes of 32 nodes and larger, parallel simulation is more cost-effective than sequential simulation.
Simple COMA node implementations
34 Citations1994Erik Hägersten, Ashley Saulsbury +1 more
The authors introduce the idea of a simple COMA architecture, a hybrid with hardware support only for the functionality frequently used, and because of its simplicity it should be quick and cheap to develop and engineer.
ACM SIGMETRICS Performance Evaluation ReviewActive memory
28 Citations1995Alvin R. Lebeck, David A. Wood
ACM Transactions on Programming Languages and SystemsEfficiently counting program events with support for on-line queries
26 Citations1994Thomas Ball
This paper represents an instrumentation method for efficiently counting events in a program's execution, with support for on-line queries of the event count, and guarantees that accurate event counts can be obtained efficiently at every point in the execution.
ACM SIGPLAN NoticesTrap-driven simulation with Tapeworm II
26 Citations1994Richard Uhlig, David F. Nagle +2 more
Trap-driven simulation with Tapeworm II
21 Citations1994Richard Uhlig, David F. Nagle +2 more
This paper compares trap-driven simulation, as implemented in Tapeworm, with the more common technique of trace-driven memory simulation with respect to speed, accuracy, portability and flexibility.
Cost/performance of a parallel computer simulator
15 Citations1994Babak Falsafi, David A. Wood
A performance model of the Wisconsin Wind Tunnel, a system that simulates cache-coherent shared-memory machines on a message-passing Thinking Machines CM-5, indicates that for target system sizes of 32 nodes and larger, parallel simulation is more cost-effective than sequential simulation.
Active memory
10 Citations1995Alvin R. Lebeck, David A. Wood
This paper describes the active memory abstraction for memory-system simulation, designed specifically for on-the-fly simulation, where memory references logically invoke a user-specified function depending upon the reference's type and accessed memory block state.
Systematic prototyping of superscalar computer architectures
4 Citations2003Thomas M. Conte, Wen mei Hwu
It is argued that the correct solution to the computer architecture design process is a prototyping approach, where the initial design is selected using the workload itself and is an architectural prototype, specifying the high-level design decisions that are difficult to acquire via detailed simulation.