----http://wwwzenger.informatik.tu-muenchen.de/persons/ruede/vorles/wiss1/sc/skript/node71.html
Paper mentioning benchmarking of single and multiprocessor machines,
appears to be part of a larger multi-chaptered document. Mentions Linpack,
SPEC, Dhrystone, etc.
----http://www.specbench.org
SPEC's webpage
----USENET: comp.benchmarks
----http://sacam.oren.ortn.edu/~dave/benchmark-faq.html
Usenet group comp.benchmarks FAQ
----http://www.csrd.uiuc.edu/benchmark/benchmark.html
Performance Evaluation and Benchmarking Projects at CSRD
----http://reality.sgi.com/employees/lm/lmbench/lmbench.html
What is lmbench?
Suite of simple, portable benchmarks
Compares different systems performance
Results available for most major vendors (SUN, HP, IBM, DEC,
SGI, PCs including 200 Mhz P6's)
Free software, covered by the GNU General Public License.
----http://mosquitonet.stanford.edu/%7Elaik/benchmarks/cached-nfs-results.html
NFS performance results (using Bonnie) of Linux vs. FreeBSD
----http://bugle.cs.uiuc.edu/
PABLO group at the University of Illinois. Developing tools for
evaluating performance of massively parallel systems. Several ongoing projects,
including:
Pablo Performance Analysis Environment
High-Performance Fortran Integration (HPF)
Virtual Reality for Performance Data Immersion
WWW Server Performance Analysis
Portable Parallel File System (PPFS)
Input/Output Characterization
Scalable I/O Initiative
Also has a page discussing the philosophy of building such tools, including a
white paper and slides.
-----http://heureka.inf.elte.hu/%7Ecsepregi/motif/perf.html
A paper exploring issues with XWINDOWS performance
-----http://www.ito.darpa.mil/ResearchAreas/Networking_Systems/High_Performance_Networking.html
Information about high performance networking machines (switchers,
routers, brouters, etc...
-----http://www.ncsa.uiuc.edu/Apps/perf/NCSAperf_homepage.html
Software performance at NCSA
-----http://wwwcn.cern.ch/pdp/pa/aquila/aquila.html
This report summarises the results of the CERN benchmark tests carried out
on a variety of Mainfames and Workstations during the last fifteen years. The
tests are a suite of FORTRAN programs used to determine the CPU power of a
computer system for running High Energy Physics applications. They are
essentially scalar due to the well known difficulties in vectorising this type
of application, but a matrix inversion in FORTRAN is used to determine the
vector potential of the system under test. This can be important for certain
applications such as accelerator design, structural analysis, or lattice gauge
calculations. The results from the early tests are normalised in terms of an
IBM 370 Series, Model 168 and later in terms of a DEC VAX 8600 which is
considered to have roughly the same capacity for HEP applications (1 CERN Unit).
After a description of both the old and new tests, the results are presented in
tabular form.
-----http://www.cs.virginia.edu/stream/
Department of Computer Science
School of Engineering and Applied Science
University of Virginia, Charlottesville, Virginia
STREAM:
Measuring Sustainable Memory Bandwidth in High Performance Computers
-----http://www.sissa.it/furio/mdbnch.html
MDBENCH, a molecular dynamics benchmark
-----http://www.dfw.net/~sdw/index.html
System Optimization Page dealing with PC type computers and benchmark
comparisons of different processors, along withj upgrade tips, etc.
-----http://web1.zdnet.com/zdlabs/ or http://web1.zdnet.com/zdbop/
Ziff-Davis Publishing Benchmarks page, claiming to be "The Ultimate
Benchmark". The second link is the actual downloadable source code for
the benchmarks.
-----http://www.silkroad.com/linux-bm.html
The Linux Benchmarks
.