Debian Codegraph @

doc/references-lmbench3

raw
1 %z Article
2 %K Staelin98
3 %A Carl Staelin
4 %A Larry McVoy
5 %T mhz: Anatomy of a microbenchmark
6 %B Proceedings USENIX Annual Technical Conference
7 %C New Orleans, LA
8 %D June 1998
9 %P 155-166
10
11 %z Article
12 %K McVoy96
13 %A Larry McVoy
14 %A Carl Staelin
15 %T lmbench: Portable tools for performance analysis
16 %B Proceedings USENIX Winter Conference
17 %C San Diego, CA
18 %D January 1996
19 %P 279-284
20
21 %K Bray90
22 %A Tim Bray
23 %T Bonnie benchmark
24 %D 1990
25 %o http://www.textuality.com/bonnie/
26
27 %z Article
28 %K Brown97
29 %A Aaron Brown
30 %A Margo Seltzer
31 %T Operating system benchmarking in the wake of lmbench: a case study of the performance of NetBSD on the Intel x86 architecture
32 %B Proceedings of the 1997 ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems
33 %C Seattle, WA
34 %D June 1997
35 %P 214-224
36 %o http://www.eecs.harvard.edu/~vino/perf/hbench/sigmetrics/hbench.html
37
38 %z Article
39 %A Cristina Hristea
40 %A Danial Lenoski
41 %A John Keen
42 %T Measuring memory hierarchy performance of cache-coherent multiprocessors using microbenchmarks
43 %B Proceedings of Supercomputing '97
44 %D November 1997
45 %C San Jose, CA
46 %o http://www.supercomp.org/sc97/proceedings/TECH/HRISTEA/
47
48 %z Thesis
49 %K Prestor01
50 %A Uros Prestor
51 %T Evaluating the memory performance of a ccNUMA system
52 %I Department of Computer Science, University of Utah
53 %D May 2001
54
55 %z Thesis
56 %K Saavedra92
57 %A Rafael H. Saavedra-Barrera
58 %T CPU Performance evaluation and execution time prediction using narrow spectrum benchmarking
59 %I Department of Computer Science, University of California at Berkeley
60 %D 1992
61
62 %z Article
63 %K Saavedra95
64 %A R.H. Saavedra
65 %A A.J. Smith
66 %T Measuring cache and TLB performance and their effect on benchmark runtimes
67 %J IEEE Transactions on Computers
68 %V 44
69 %N 10
70 %D October 1995
71 %P 1223-1235
72
73 %z Article
74 %K Wolman89
75 %A Barry L. Wolman
76 %A Thomas M. Olson
77 %T IOBENCH: a system independent IO benchmark
78 %J Computer Architecture News
79 %V 17
80 %N 5
81 %D September 1989
82 %P 55-70
83 %x IOBENCH is an operating system and processor independent synthetic
84 %x input/output (IO) benchmark designed to put a configurable IO and
85 %x processor (CP) load on the system under test. This paper discusses
86 %x the UNIX versions.
87 %k IOBENCH, synthetic I/O benchmark, UNIX workload
88 %s vinton%[email protected] (Fri Sep 20 12:55:58 PDT 1991)
89
90 %z Book
91 %K Hennessy96
92 %A John L. Hennessy
93 %A David A. Patterson
94 %T Computer Architecture A Quantitative Approach, 2nd Edition
95 %I Morgan Kaufman
96 %D 1996
97
98 %z Book
99 %K Jain91
100 %A Raj Jain
101 %T The Art of Computer Systems Performance Analysis: Techniques for Experimental Design, Measurement, Simulation, and Modeling
102 %I Wiley-Interscience
103 %C New York, NY
104 %D April 1991
105
106 %z Article
107 %K Chen94a
108 %A P. M. Chen
109 %A D. A. Patterson
110 %T A new approach to I/O performance evaluation \- self-scaling I/O benchmarks, predicted I/O performance
111 %D November 1994
112 %J Transactions on Computer Systems
113 %V 12
114 %N 4
115 %P 308-339
116 %x Current I/O benchmarks suffer from several chronic problems: they
117 %x quickly become obsolete; they do not stress the I/O system; and they
118 %x do not help much in undelsi;anding I/O system performance. We
119 %x propose a new approach to I/O performance analysis. First, we
120 %x propose a self-scaling benchmark that dynamically adjusts aspects of
121 %x its workload according to the performance characteristic of the
122 %x system being measured. By doing so, the benchmark automatically
123 %x scales across current and future systems. The evaluation aids in
124 %x understanding system performance by reporting how performance varies
125 %x according to each of five workload parameters. Second, we propose
126 %x predicted performance, a technique for using the results from the
127 %x self-scaling evaluation to estimate quickly the performance for
128 %x workloads that have not been measured. We show that this technique
129 %x yields reasonably accurate performance estimates and argue that this
130 %x method gives a far more accurate comparative performance evaluation
131 %x than traditional single-point benchmarks. We apply our new
132 %x evaluation technique by measuring a SPARCstation 1+ with one SCSI
133 %x disk, an HP 730 with one SCSI-II disk, a DECstation 5000/200 running
134 %x the Sprite LFS operating system with a three-disk disk array, a
135 %x Convex C240 minisupercomputer with a four-disk disk array, and a
136 %x Solbourne 5E/905 fileserver with a two-disk disk array.
137 %s [email protected] (Mon Mar 13 10:57:38 1995)
138 %s wilkes%[email protected] (Sun Mar 19 12:38:01 PST 1995)
139 %s wilkes%[email protected] (Sun Mar 19 12:38:53 PST 1995)
140
141 %z InProceedings
142 %K Ousterhout90
143 %s wilkes%[email protected] (Fri Jun 29 20:46:08 PDT 1990)
144 %A John K. Ousterhout
145 %T Why aren't operating systems getting faster as fast as hardware?
146 %B Proceedings USENIX Summer Conference
147 %C Anaheim, CA
148 %D June 1990
149 %P 247-256
150 %x This paper evaluates several hardware pplatforms and operating systems using
151 %x a set of benchmarks that stress kernel entry/exit, file systems, and
152 %x other things related to operating systems. The overall conclusion is that
153 %x operating system performance is not improving at the same rate as the base speed of the
154 %x underlying hardware. The most obvious ways to remedy this situation
155 %x are to improve memory bandwidth and reduce operating systems'
156 %x tendency to wait for disk operations to complete.
157 %o Typical performance of 10-20 MIPS cpus is only 0.4 times what
158 %o their raw hardware performance would suggest. HP-UX is
159 %o particularly bad on the HP 9000/835, at about 0.2x. (Although
160 %o this measurement discounted a highly-tuned getpid call.)
161 %k OS performance, RISC machines, HP9000 Series 835 system calls
162
163 %z InProceedings
164 %K McVoy91
165 %A L. W. McVoy
166 %A S. R. Kleiman
167 %T Extent-like Performance from a Unix File System
168 %B Proceedings USENIX Winter Conference
169 %C Dallas, TX
170 %D January 1991
171 %P 33-43
172
173 %z Article
174 %K Chen93d
175 %A Peter M. Chen
176 %A David Patterson
177 %T Storage performance \- metrics and benchmarks
178 %J Proceedings of the IEEE
179 %V 81
180 %N 8
181 %D August 1993
182 %P 1151-1165
183 %x Discusses metrics and benchmarks used in storage performance evaluation.
184 %x Describes, reviews, and runs popular I/O benchmarks on three systems. Also
185 %x describes two new approaches to storage benchmarks: LADDIS and a Self-Scaling
186 %x benchmark with predicted performance.
187 %k I/O, storage, benchmark, workload, self-scaling benchmark,
188 %k predicted performance, disk, performance evaluation
189 %s staelin%[email protected] (Wed Sep 27 16:21:11 PDT 1995)
190
191 %z Article
192 %K Park90a
193 %A Arvin Park
194 %A J. C. Becker
195 %T IOStone: a synthetic file system benchmark
196 %J Computer Architecture News
197 %V 18
198 %N 2
199 %D June 1990
200 %P 45-52
201 %o this benchmark is useless for all modern systems; it fits
202 %o completely inside the file system buffer cache. Soon it may even
203 %o fit inside the processor cache!
204 %k IOStone, I/O, benchmarks
205 %s staelin%[email protected] (Wed Sep 27 16:37:26 PDT 1995)
206
207 %z Article
208 %K Fenwick95
209 %A David M. Fenwick
210 %A Denis J. Foley
211 %A William B. Gist
212 %A Stephen R. VanDoren
213 %A Danial Wissell
214 %T The AlphaServer 8000 series: high-end server platform development
215 %J Digital Technical Journal
216 %V 7
217 %N 1
218 %D August 1995
219 %P 43-65
220 %x The AlphaServer 8400 and the AlphaServer 8200 are Digital's newest high-end
221 %x server products. Both servers are based on the 300Mhz Alpha 21164
222 %x microprocessor and on the AlphaServer 8000-series platform architecture.
223 %x The AlphaServer 8000 platform development team set aggressive system data
224 %x bandwidth and memory read latency targets in order to achieve high-performance
225 %x goals. The low-latency criterion was factored into design decisions made at
226 %x each of the seven layers of platform development. The combination of
227 %x industry-leading microprocessor technology and a system platform focused
228 %x on low latency has resulted in a 12-processor server implementation ---
229 %x the AlphaServer 8400 --- capable of supercomputer levels of performance.
230 %k DEC Alpha server, performance, memory latency
231 %s staelin%[email protected] (Wed Sep 27 17:27:23 PDT 1995)
232
233 %z Book
234 %K Toshiba94
235 %Q Toshiba
236 %T DRAM Components and Modules
237 %I Toshiba America Electronic Components, Inc.
238 %P A59-A77,C37-C42
239 %D 1994
240
241 %z Article
242 %K McCalpin95
243 %A John D. McCalpin
244 %T Memory bandwidth and machine balance in current high performance computers
245 %J IEEE Technical Committee on Computer Architecture newsletter
246 %D December 1995
247
248 %z Article
249 %K McCalpin02
250 %A John D. McCalpin
251 %T The STREAM2 home page
252 %o http://www.cs.virginia.edu/stream/stream2/
253 %D 2002
254
255 %z Article
256 %K FSF89
257 %A Richard Stallman
258 %Q Free Software Foundation
259 %T General Public License
260 %D 1989
261 %O Included with \*[lmbench]
262
263 %z Article
264 %K Shein89
265 %A Barry Shein
266 %A Mike Callahan
267 %A Paul Woodbury
268 %T NFSSTONE: A network file server performance benchmark
269 %B Proceedings USENIX Summer Conference
270 %C Baltimore, MD
271 %D June 1989
272 %P 269-275
273
274 %z Article
275 %K Weicker84
276 %A R.P. Weicker
277 %T Dhrystone: A synthetic systems programming benchmark
278 %J Communications of the ACM
279 %V 27
280 %N 10
281 %P 1013--1030
282 %D 1984
283
284 %z Article
285 %K Howard88
286 %A J. Howard
287 %A M. Kazar
288 %A S. Menees
289 %A S. Nichols
290 %A M. Satyanrayanan
291 %A R. Sidebotham
292 %A M. West
293 %T Scale and performance in a distributed system
294 %J ACM Transactions on Computer Systems
295 %V 6
296 %N 1
297 %D February 1988
298 %P 51--81
299 %k Andrew benchmark
300
301 %z Article
302 %K Banga97
303 %A Guarav Banga
304 %A Peter Druschel
305 %T Measuring the capacity of a web server
306 %B Proceedings USENIX Symposium on Internet Technologies and Systems
307 %C Monterey, CA
308 %D December 1997
309 %P 61--71
310
311 %z Article
312 %K Banga98
313 %A Guarav Banga
314 %A Jeffrey C. Mogul
315 %T Scalable kernel performance for internet servers under realistic loads
316 %B Proceedings of the 1998 USENIX Annual Technical Conference
317 %C New Orleans, LA
318 %D June 1998
319 %P 69--83
320
321 %z Article
322 %K Mogul99
323 %A Jeffrey C. Mogul
324 %T Brittle metrics in operating systems research
325 %B Proceedings 7th IEEE Workshop on Hot Topics in Operating Systems (HotOS-VII)
326 %C Rio Rico, AZ
327 %P 90--95
328 %D March 1999
329
330 %z Article
331 %K Regehr2002
332 %A John Regehr
333 %T Inferring scheduling behavior with Hourglass
334 %B Proceedings of the USENIX Annual Technical Conference FREENIX track
335 %C Monterey, CA
336 %D June 2002
337 %P 143--156
338
339 %z Article
340 %K Seltzer99
341 %A Margo Seltzer
342 %A David Krinsky
343 %A Keith Smith
344 %A Xiolan Zhang
345 %T The case for application-specific benchmarking
346 %B Proceedings of the 1999 Workshop on Hot Topics in Operating Systems
347 %C Rico, AZ
348 %D 1999
349 %P 102--107
350
351 %z Article
352 %K Smith97
353 %A Keith A. Smith
354 %A Margo L. Seltzer
355 %T File system aging --- Increasing the relevance of file system benchmarks
356 %B Proceedings of the 1997 SIGMETRICS Conference
357 %D June 1997
358 %C Seattle, WA
359 %P 203-213
360
361 %z Article
362 %K Tullsen96
363 %A Dean Tullsen
364 %A Susan Eggers
365 %A Joel Emer
366 %A Henry Levy
367 %A Jack Lo
368 %A Rebecca Stamm
369 %T Exploiting choice: Instruction fetch and issue on an implementable simultaneous multithreading processor
370 %C Proceedings of the 23rd Annual International Symposium on Computer Architecture
371 %D May 1996
372 %P 191-202
373 %O http://www.cs.washington.edu/research/smt/papers/ISCA96.ps
374
375 %z Article
376 %K Tullsen99
377 %A Dean Tullsen
378 %A Jack Lo
379 %A Susan Eggers
380 %A Henry Levy
381 %T Supporting fine-grain synchronization on a simultaneous multithreaded processor
382 %B Proceedings of the 5th International Symposium on High Performance Computer Architecture
383 %D January 1999
384 %P 54-58
385 %O http://www.cs.washington.edu/research/smt/papers/hpca.ps
386
387 %z Report
388 %K Whaley97
389 %A R. Clint Whaley
390 %A Jack Dongarra
391 %T Automatically tuned linear algebra software
392 %I Department of Computer Science, University of Tennessee
393 %C Knoxville, TN
394 %R UT-CS-97-366
395 %D 1997
396 %o http://math-atlas.sourceforge.net/
397
398 %z Article
399 %K SPEChpc96
400 %Q Standard Performance Evaluation Corporation
401 %T SPEC HPC96 benchmark
402 %D 1996
403 %O http://www.specbench.org/hpg/hpc96/
404
405 %z Article
406 %K Parkbench
407 %Q PARallel Kernels and BENCHmarks committee
408 %T PARKBENCH
409 %D 2002
410 %O http://www.netlib.org/parkbench/
411
412 %z Article
413 %K NAS
414 %Q NASA Advanced Supercomputing Division, NASA Ames Research Center
415 %T NAS parallel benchmarks
416 %O http://www.nas.nasa.gov/NAS/NPB
417
418 %z Article
419 %K Glendinning94
420 %A Ian Glendinning
421 %T GENESIS distributed memory benchmark suite
422 %O http://wotug.ukc.ac.uk/parallel/performance/benchmarks/genesis
423 %D 1994
424
425 %z Article
426 %K Intel99
427 %Q Intel
428 %T Profusion --- An 8-way symmetric multiprocessing chipset
429 %O http://netserver.hp.com/docs/download.asp?file=tp_profusion(r).pdf
430 %D July 1999