Lines Matching refs:netperf

1 This is netperf.info, produced by makeinfo version 4.13 from
2 netperf.texi.
5 netperf benchmark.
10 document per the terms of the netperf source license, a copy of
11 which can be found in the file `COPYING' of the basic netperf
15 File: netperf.info, Node: Top, Next: Introduction, Prev: (dir), Up: (dir)
21 netperf benchmark.
26 document per the terms of the netperf source license, a copy of
27 which can be found in the file `COPYING' of the basic netperf
32 * Introduction:: An introduction to netperf - what it
34 * Installing Netperf:: How to go about installing netperf.
50 File: netperf.info, Node: Introduction, Next: Installing Netperf, Prev: Top, Up: Top
75 While not every revision of netperf will work on every platform
76 listed, the intention is that at least some version of netperf will
91 here. While it is often used by them, netperf is NOT supported via any
93 to make enhancements and modifications to netperf to suit your
95 netperf copyright. If you feel so inclined, you can send your changes
96 to netperf-feedback <netperf-feedback@netperf.org> for possible
97 inclusion into subsequent versions of netperf.
99 It is the Contributing Editor's belief that the netperf license walks
106 The netperf-talk <netperf-talk@netperf.org> mailing list is
107 available to discuss the care and feeding of netperf with others who
109 netperf-talk mailing list is a closed list (to deal with spam) and you
110 must first subscribe by sending email to netperf-talk-request
111 <netperf-talk-request@netperf.org>.
118 File: netperf.info, Node: Conventions, Prev: Introduction, Up: Introduction
125 netperf parameters. If you wish to set both parameters to separate
170 netperf <global> -- <test-specific>
173 File: netperf.info, Node: Installing Netperf, Next: The Design of Netperf, Prev: Introduction, …
181 styles of netperf installation. The first runs the netperf server
208 File: netperf.info, Node: Getting Netperf Bits, Next: Installing Netperf Bits, Prev: Installing …
213 Gzipped tar files of netperf sources can be retrieved via anonymous FTP
214 (ftp://ftp.netperf.org/netperf) for "released" versions of the bits.
216 from the experimental (ftp://ftp.netperf.org/netperf/experimental)
220 is provided via the NetperfPage (http://www.netperf.org/)
222 The bits corresponding to each discrete release of netperf are
223 tagged (http://www.netperf.org/svn/netperf2/tags) for retrieval via
225 corresponding to this version of the manual - netperf 2.6.0
226 (http://www.netperf.org/svn/netperf2/tags/netperf-2.6.0). Those
227 wishing to be on the bleeding edge of netperf development can use
229 (http://www.netperf.org/svn/netperf2/trunk). When fixing bugs or
233 download netperf bits. These may be simple mirrors of the main netperf
234 site, or they may be local variants on netperf. As with anything one
239 As a general rule, binaries of netperf and netserver are not
240 distributed from ftp.netperf.org. From time to time a kind soul or
241 souls has packaged netperf as a Debian package available via the
246 File: netperf.info, Node: Installing Netperf Bits, Next: Verifying Installation, Prev: Getting N…
251 Once you have downloaded the tar file of netperf sources onto your
252 system(s), it is necessary to unpack the tar file, cd to the netperf
256 gzcat netperf-<version>.tar.gz | tar xf -
257 cd netperf-<version>
269 If the netperf configure script does not know how to automagically
273 that area, feel free to contact <netperf-feedback@netperf.org>.
281 Starting with version 2.5.0, netperf began migrating most of the
282 "classic" netperf tests found in `src/nettest_bsd.c' to the so-called
290 attempt to obtain resolution via <netperf-talk@netperf.org> or
291 <netperf-feedback@netperf.org>. If that is unsuccessful, you can add a
295 Starting with version 2.5.0, netperf includes the "burst mode"
298 <netperf-talk@netperf.org> or <netperf-feedback@netperf.org>. If that
303 command with a CFLAGS and/or LIBS variable as the netperf configure
310 `--enable-intervals=yes' to give netperf the ability to "pace" its
311 _STREAM tests and `--enable-histogram=yes' to have netperf keep a
315 measurable. For example, the histogram of a netperf TCP_RR test will
317 netperf -t TCP_RR -H lag -v 2
352 $ src/netperf -D 1.35 -H tardy.hpl.hp.com -f M
376 Those familiar with this option in earlier versions of netperf will
391 effect an automagic editing of the necessary files to have netperf
402 If the compilation of netperf or netserver happens to fail, feel free
403 to contact <netperf-feedback@netperf.org> or join and ask in
404 <netperf-talk@netperf.org>. However, it is quite important that you
410 File: netperf.info, Node: Verifying Installation, Prev: Installing Netperf Bits, Up: Installing …
415 Basically, once netperf is installed and netserver is configured as a
417 netperf
419 $ netperf
429 File: netperf.info, Node: The Design of Netperf, Next: Global Command-line Options, Prev: Instal…
435 executables - netperf and netserver. Generally you will only execute
436 the netperf program, with the netserver program being invoked by the
440 When you execute netperf it will establish a "control connection" to
452 control connection and combined with netperf's result for display to
466 File: netperf.info, Node: CPU Utilization, Prev: The Design of Netperf, Up: The Design of Netperf
478 CPU utilization in netperf is reported as a value between 0 and 100%
480 utilization, netperf will report a metric called a "service demand".
494 indicate which mechanism was used on each of the local (netperf) and
500 The CPU utilization measurement mechanism was unknown to netperf or
501 netperf/netserver was not compiled to include CPU utilization
548 netperf goes through some hand-waving to try to "fix" the problem.
587 Other mechanisms included in netperf in the past have included
595 benchmarks. These mechanisms are not available in netperf 2.4.0
605 when the system is running netperf and the ratio is used to compute CPU
608 In the past, netperf included some mechanisms that only reported CPU
610 removed from netperf versions 2.4.0 and later because they are
630 mpstat are often based on the same mechanisms used by netperf.
637 File: netperf.info, Node: CPU Utilization in a Virtual Guest, Prev: CPU Utilization, Up: CPU Uti…
642 The CPU utilization mechanisms used by netperf are "inline" in that
643 they are run by the same netperf or netserver process as is running the
655 should not really trust CPU utilization figures reported by netperf or
660 netperf _RR tests - path-lengths and overheads can be a significant
667 File: netperf.info, Node: Global Command-line Options, Next: Using Netperf to Measure Bulk Data T…
673 available in the netperf and netserver binaries. Essentially, it is an
674 expanded version of the usage information displayed by netperf or
683 File: netperf.info, Node: Command-line Options Syntax, Next: Global Options, Prev: Global Comman…
688 Revision 1.8 of netperf introduced enough new functionality to overrun
696 affect nearly any and every test type of netperf. The second type are
704 netperf <global> -- <test-specific>
706 netperf <global>
708 netperf -- <test-specific>
712 File: netperf.info, Node: Global Options, Prev: Command-line Options Syntax, Up: Global Command-…
734 This option is only present when netperf has been configure with
742 -P) output of netperf.
764 sent by default to the file `/tmp/netperf.debug'. [Default: no
768 This option is only available when netperf is configured with
769 -enable-demo=yes. When set, it will cause netperf to emit periodic
798 of netperf removed that restriction, creating a temporary file as
803 This option causes netperf to display its "global" usage string and
811 netperf to use IPv4 addressing only.
848 asks netperf to be 99% confident that the measured mean values for
854 If classic netperf test calculates that the desired confidence
858 netperf -H tardy.cup -i 3 -I 99,5
875 In the example above we see that netperf did not meet the desired
883 intervals will not result in netperf emitting a warning. To
888 intervals will remain in a "migrated" classic netperf test.
900 migrated netperf test will not display the actual number of
917 This option instructs netperf to keep additional timing statistics
951 will be microseconds. Added in netperf 2.5.0.
971 total length of the netperf test will be somewhere between the
978 useful when you wish to make sure that the netperf control and data
980 one is trying to run netperf through those evil, end-to-end
987 This option tells netperf how many CPUs it should ass-u-me are
988 active on the system running netperf. In particular, this is used
990 calculations. On certain systems, netperf is able to determine
992 number netperf might be able to determine on its own.
995 system running netserver. When netperf/netserver cannot
999 As it is almost universally possible for netperf/netserver to
1002 in a future release of netperf.
1005 This option tells netperf to forgo establishing a control
1006 connection. This makes it is possible to run some limited netperf
1025 socket buffer size and the like via the netperf command line. Nor
1038 Added in netperf 2.4.3.
1045 will cause the buffers passed to the local (netperf) send and
1056 netperf the port number at which it should expect the remote
1058 value of the optionspec will request netperf to bind to that local
1062 tells netperf that the remote netserver is listening on port 12345
1067 causes netperf to bind to the local port number 32109 before
1071 one is looking to run netperf through those evil, end-to-end
1084 This option will cause netperf to sleep `<seconds>' before
1086 useful in situations where one wishes to start a great many netperf
1095 attempt will be made on the netperf side of all tests, and will be
1103 This option is used to tell netperf which test you wish to run.
1124 Not all tests are always compiled into netperf. In particular, the
1126 netperf when configured with
1136 affinity of netperf and/or netserver.
1137 netperf -T 1
1138 will bind both netperf and netserver to "CPU 1" on their respective
1140 netperf -T 1,
1141 will bind just netperf to "CPU 1" and will leave netserver unbound.
1142 netperf -T ,2
1143 will leave netperf unbound and will bind netserver to "CPU 2."
1144 netperf -T 1,2
1145 will bind netperf to "CPU 1" and netserver to "CPU 2."
1152 This option controls how verbose netperf will be in its output,
1163 If the verbosity level is set to "1" then the "normal" netperf
1170 call or for each transaction if netperf was configured with
1180 This option displays the netperf version and then exits.
1182 Added in netperf 2.4.4.
1185 If netperf was configured with `--enable-intervals=yes' then this
1194 Unlike some benchmarks, netperf does not continuously send or
1219 File: netperf.info, Node: Using Netperf to Measure Bulk Data Transfer, Next: Using Netperf to Mea…
1227 network. The classic netperf convention for a bulk data transfer test
1236 File: netperf.info, Node: Issues in Bulk Transfer, Next: Options common to TCP UDP and SCTP tests…
1256 A bulk-transfer test can be CPU bound even when netperf reports less
1260 of a netperf _STREAM test cannot make use of much more than the power
1261 of one CPU. Exceptions to this generally occur when netperf and/or
1294 netperf was running. In such situations something along the lines of:
1296 netperf -t TCP_mumble...
1313 output selction option and netperf will report the retransmissions
1326 File: netperf.info, Node: Options common to TCP UDP and SCTP tests, Prev: Issues in Bulk Transfer…
1390 This option sets the local (netperf) send and receive socket buffer
1461 File: netperf.info, Node: TCP_STREAM, Next: TCP_MAERTS, Prev: Options common to TCP UDP and SCTP…
1466 The TCP_STREAM test is the default test in netperf. It is quite
1468 netperf to the system running netserver. While time spent establishing
1471 test is. This is how netperf knows that all the data it sent was
1484 present netperf does not perform any explicit flush operations.
1491 netperf does not validate command-line options to enforce that.
1519 $ netperf -H lag
1540 File: netperf.info, Node: TCP_MAERTS, Next: TCP_SENDFILE, Prev: TCP_STREAM, Up: Options common …
1547 netperf. The global command-line `-F' option is ignored for this test
1555 $ netperf -H lag -t TCP_MAERTS -- -s 128K -S 128K
1571 File: netperf.info, Node: TCP_SENDFILE, Next: UDP_STREAM, Prev: TCP_MAERTS, Up: Options common …
1577 netperf the platform's `sendfile()' call instead of calling `send()'.
1595 specified, netperf will create a temporary file, which it will delete
1603 $ netperf -H lag -F ../src/netperf -t TCP_SENDFILE -- -s 128K -S 128K
1610 $ netperf -H lag -F /boot/vmlinuz-2.6.8-1-686 -t TCP_SENDFILE -- -s 128K -S 128K
1622 File: netperf.info, Node: UDP_STREAM, Next: XTI_TCP_STREAM, Prev: TCP_SENDFILE, Up: Options com…
1631 and neither does netperf. However, if you wish, you can configure
1632 netperf with `--enable-intervals=yes' to enable the global command-line
1647 $ netperf -t UDP_STREAM -H 192.168.2.125 -- -m 32768
1656 The first line of numbers are statistics from the sending (netperf)
1662 buffer size (`-s' option) netperf will likely abort with an error
1665 netperf -t UDP_STREAM -H 192.168.2.125
1674 $ netperf -t UDP_STREAM -H 192.168.2.125 -- -m 65000 -S 32768
1692 File: netperf.info, Node: XTI_TCP_STREAM, Next: XTI_UDP_STREAM, Prev: UDP_STREAM, Up: Options c…
1700 device files, which is required by the `t_open()' call made by netperf
1703 The XTI_TCP_STREAM test is only present if netperf was configured
1708 File: netperf.info, Node: XTI_UDP_STREAM, Next: SCTP_STREAM, Prev: XTI_TCP_STREAM, Up: Options …
1716 device files, which is required by the `t_open()' call made by netperf
1719 The XTI_UDP_STREAM test is only present if netperf was configured
1724 File: netperf.info, Node: SCTP_STREAM, Next: DLCO_STREAM, Prev: XTI_UDP_STREAM, Up: Options com…
1736 The SCTP_STREAM test is only present if netperf was configured with
1741 File: netperf.info, Node: DLCO_STREAM, Next: DLCL_STREAM, Prev: SCTP_STREAM, Up: Options common…
1785 The DLCO_STREAM test is only present if netperf was configured with
1790 File: netperf.info, Node: DLCL_STREAM, Next: STREAM_STREAM, Prev: DLCO_STREAM, Up: Options comm…
1800 and reassembly available and netperf does not presume to provide one.
1805 The DLCL_STREAM test is only present if netperf was configured with
1810 File: netperf.info, Node: STREAM_STREAM, Next: DG_STREAM, Prev: DLCL_STREAM, Up: Options common…
1824 The STREAM_STREAM test is only present if netperf was configured with
1829 File: netperf.info, Node: DG_STREAM, Prev: STREAM_STREAM, Up: Options common to TCP UDP and SCTP…
1842 The DG_STREAM test is only present if netperf was configured with
1847 File: netperf.info, Node: Using Netperf to Measure Request/Response, Next: Using Netperf to Measu…
1861 a time test such as those found by default in netperf, is particularly
1868 netperf _RR test.
1877 2.5.0 (actually slightly before) netperf still does not do the latter,
1879 netperf test, or includes the appropriate *note output selector: Omni
1890 File: netperf.info, Node: Issues in Request/Response, Next: Options Common to TCP UDP and SCTP _R…
1897 important as netperf generally only has one transaction outstanding at
1905 particularly if one CPU binds the netperf and netserver differently via
1936 File: netperf.info, Node: Options Common to TCP UDP and SCTP _RR tests, Prev: Issues in Request/R…
1984 This option sets the local (netperf) send and receive socket buffer
1993 Will request the local send (netperf) and receive socket buffer
2049 File: netperf.info, Node: TCP_RR, Next: TCP_CC, Prev: Options Common to TCP UDP and SCTP _RR tes…
2085 netperf -t TCP_RR -H 192.168.2.125
2101 File: netperf.info, Node: TCP_CC, Next: TCP_CRR, Prev: TCP_RR, Up: Options Common to TCP UDP an…
2127 this happening, netperf will by default select its own client port
2130 second. The size of the client port space used by netperf can be
2133 port numbers used by netperf at the client end.
2147 File: netperf.info, Node: TCP_CRR, Next: UDP_RR, Prev: TCP_CC, Up: Options Common to TCP UDP an…
2159 test was added to netperf to simulate just that.
2170 File: netperf.info, Node: UDP_RR, Next: XTI_TCP_RR, Prev: TCP_CRR, Up: Options Common to TCP UD…
2180 netperf does not add anything for that either. This means that if
2190 The netperf side of a UDP_RR test will call `connect()' on its data
2200 $ netperf -T 1 -H 192.168.1.213 -t UDP_RR -c -C
2212 option was used to make sure netperf and netserver ran on a given CPU
2216 File: netperf.info, Node: XTI_TCP_RR, Next: XTI_TCP_CC, Prev: UDP_RR, Up: Options Common to TCP…
2230 File: netperf.info, Node: XTI_TCP_CC, Next: XTI_TCP_CRR, Prev: XTI_TCP_RR, Up: Options Common t…
2243 File: netperf.info, Node: XTI_TCP_CRR, Next: XTI_UDP_RR, Prev: XTI_TCP_CC, Up: Options Common t…
2256 File: netperf.info, Node: XTI_UDP_RR, Next: DLCL_RR, Prev: XTI_TCP_CRR, Up: Options Common to T…
2270 File: netperf.info, Node: DLCL_RR, Next: DLCO_RR, Prev: XTI_UDP_RR, Up: Options Common to TCP U…
2276 File: netperf.info, Node: DLCO_RR, Next: SCTP_RR, Prev: DLCL_RR, Up: Options Common to TCP UDP …
2282 File: netperf.info, Node: SCTP_RR, Prev: DLCO_RR, Up: Options Common to TCP UDP and SCTP _RR tes…
2288 File: netperf.info, Node: Using Netperf to Measure Aggregate Performance, Next: Using Netperf to …
2294 use when one wants to measure aggregate performance because netperf has
2299 There are a few ways to measure aggregate performance with netperf.
2300 The first is to run multiple, concurrent netperf tests and can be
2301 applied to any of the netperf tests. The second is to configure
2302 netperf with `--enable-burst' and is applicable to the TCP_RR test. The
2312 File: netperf.info, Node: Running Concurrent Netperf Tests, Next: Using --enable-burst, Prev: Us…
2318 wants to measure aggregate performance because netperf has no support
2323 necessary to run netperf. The skew error can be minimized by making use
2329 netperf -t TCP_STREAM -H tardy.cup.hp.com -i 10 -P 0 &
2334 concurrent netperf will iterate 10 times thanks to the `-i' option and
2350 netperf -t TCP_STREAM -H $hosts -i 10 -P 0 &
2355 If you see warnings about netperf not achieving the confidence
2364 netperf -t TCP_STREAM -H tardy.cup.hp.com -i 10 -P 0 -c -C &
2384 NOTE: It is very important to remember that netperf is calculating
2386 (those last two columns in the output above) each netperf assumes
2388 concurrent tests the service demands reported by netperf will be
2397 netperf -t TCP_STREAM -H tardy.cup.hp.com -B "this is test $i" -i 10 -P 0 &
2416 File: netperf.info, Node: Issues in Running Concurrent Tests, Prev: Running Concurrent Netperf Te…
2422 other issues to consider when running concurrent netperf tests.
2437 remember that each concurrent netperf instance is calculating
2439 netperf assumes it is the only thing running on the system. This means
2440 that for concurrent tests the service demands reported by netperf will
2447 than netperf commands. This introduces more skew error, which the
2449 possibility is to actually run three consecutive netperf tests on each
2452 started is less than the length of the first netperf command in the
2457 File: netperf.info, Node: Using --enable-burst, Next: Using --enable-demo, Prev: Running Concurr…
2467 To have burst-mode functionality present in netperf. This enables a
2471 Normally, netperf will attempt to ramp-up the number of outstanding
2475 smaller at the time than what netperf wants to have outstanding at one
2483 just a single instance of netperf, increasing the burst size can
2489 netperf -v 0 -t TCP_RR -B "-b $b" -H hpcpc108 -P 0 -- -b $b
2517 netperf -v 0 -t TCP_RR -B "-b $b -D" -H hpcpc108 -P 0 -- -b $b -D
2539 netperf tests. This would then be an "aggregate of aggregates" if you
2545 netperf -H hpcpc108 -v 0 -P 0 -i 10 -B "aggregate $i -b 8 -D" -t TCP_RR -- -b 8 -D &
2553 Since each netperf did hit the confidence intervals, we can be
2564 netperf -H 192.168.2.108 -l 60 -v 0 -P 0 -B "aggregate $i -b 8 -D" -t TCP_RR -- -b 8 -D &
2599 netperf was reporting very well.
2610 netperf -l 60 -H 192.168.2.108 -v 0 -P 0 -B "aggregate $i -b 8 -D" -t TCP_RR -- -b 8 -D & done
2642 File: netperf.info, Node: Using --enable-demo, Prev: Using --enable-burst, Up: Using Netperf to …
2649 and compile netperf to enable netperf to emit "interim results" at
2652 of netperf will then look something like
2654 $ src/netperf -D 1.25
2698 knowledge of how long it takes to get netperf tests started and/or
2703 has been nor is universally true. For that reason netperf tries to
2706 each `send' or `recv' call completes netperf tries to guess how many
2708 that many units of work have been completed will netperf check the
2709 time. If the reporting interval has passed, netperf will emit an
2710 "interim result." If the interval has not passed, netperf will update
2713 After a bit of thought one can see that if things "speed-up" netperf
2714 will still honor the interval. However, if things "slow-down" netperf
2718 netperf (not shown) is started we see things slowing down and netperf
2720 $ src/netperf -D 2 -H tardy.hpl.hp.com -l 20
2739 improve the netperf's honoring the interval but one should not ass-u-me
2745 File: netperf.info, Node: Using Netperf to Measure Bidirectional Transfer, Next: The Omni Tests, …
2750 There are two ways to use netperf to measure the performance of
2751 bidirectional transfer. The first is to run concurrent netperf tests
2752 from the command line. The second is to configure netperf with
2758 instance, using the concurrent netperf test mechanism means that
2771 File: netperf.info, Node: Bidirectional Transfer with Concurrent Tests, Next: Bidirectional Trans…
2776 If we had two hosts Fred and Ethel, we could simply run a netperf *note
2778 concurrent netperf TCP_STREAM test on Ethel pointing at Fred, but since
2779 there are no mechanisms to synchronize netperf tests and we would be
2790 netperf -H 192.168.2.108 -t TCP_STREAM -B "outbound" -i 10 -P 0 -v 0 \
2792 netperf -H 192.168.2.108 -t TCP_MAERTS -B "inbound" -i 10 -P 0 -v 0 \
2805 on which we were running netperf. Of course that sense is switched on
2816 netperf -H 192.168.1.3 -t omni -l 10 -P 0 -- \
2818 netperf -H 192.168.1.3 -t omni -l 10 -P 0 -- \
2826 File: netperf.info, Node: Bidirectional Transfer with TCP_RR, Next: Implications of Concurrent Te…
2842 netperf -f m -t TCP_RR -H 192.168.1.3 -v 2 -- -b 6 -r 32K -S 256K -S 256K
2868 netperf might attempt to do this automagically, but there are some
2872 File: netperf.info, Node: Implications of Concurrent Tests vs Burst Request/Response, Prev: Bidir…
2885 `-T' option has been used to bind netperf/netserver to specific CPUs.
2908 File: netperf.info, Node: The Omni Tests, Next: Other Netperf Tests, Prev: Using Netperf to Meas…
2913 Beginning with version 2.5.0, netperf begins a migration to the `omni'
2916 easier for netperf to support multiple protocols and report a great
2926 and remake netperf. Remaking netserver is optional because even in
2937 File: netperf.info, Node: Native Omni Tests, Next: Migrated Tests, Prev: The Omni Tests, Up: Th…
2943 the global `-t' test-selection option. This will then cause netperf to
2954 This option sets the direction of the test relative to the netperf
2959 Any of which will cause netperf to send to the netserver.
2962 Any of which will cause netserver to send to netperf.
2978 $ netperf -t omni -- -d rr -k "THROUGHPUT,THROUGHPUT_UNITS"
2990 $ netperf -t omni -- -d rr -o "THROUGHPUT,THROUGHPUT_UNITS"
3000 look quite similar to classic netperf output:
3001 $ netperf -t omni -- -d rr -O "THROUGHPUT,THROUGHPUT_UNITS"
3020 known to netperf include:
3045 This can set the send size for either or both of the netperf and
3048 sets only the netperf-side send size to 32768 bytes, and or's-in
3055 sets the netperf side send size to 16284 bytes, the netserver side
3059 This can set the receive size for either or both of the netperf and
3065 sets only the netperf side receive size to 32768 bytes and or's-in
3068 sets the netserver side receive size to 16384 bytes and the netperf
3073 File: netperf.info, Node: Migrated Tests, Next: Omni Output Selection, Prev: Native Omni Tests, …
3085 $ netperf
3116 $ netperf -t tcp_rr -- -k THROUGHPUT,THROUGHPUT_UNITS
3121 $ netperf -t tcp_rr
3132 File: netperf.info, Node: Omni Output Selection, Prev: Migrated Tests, Up: The Omni Tests
3146 identical to "classic" netperf output, is inspired by it.
3179 netperf output. The format will be the `human readable' format
3183 none is specified with the `-k', `-o' or `-O' option netperf will
3189 netperf will then assume the output specification is a filename and
3197 File: netperf.info, Node: Omni Output Selectors, Prev: Omni Output Selection, Up: Omni Output Se…
3219 This will display the data flow direction relative to the netperf
3236 This will display the local (netperf) send socket buffer size (aka
3240 This will display the local (netperf) send socket buffer size
3247 This will display the local (netperf) send socket buffer size
3255 This will display the local (netperf) receive socket buffer size
3259 This will display the local (netperf) receive socket buffer size
3266 This will display the local (netperf) receive socket buffer size
3312 This will display the size of the buffers netperf passed in any
3317 This will display the size of the buffers netperf passed in any
3332 This will display the size of the requests netperf sent in a
3341 measured by netperf. Units: 0 to 100 percent.
3345 test as measured by netperf. Only supported by netcpu_procstat.
3350 test as measured by netperf. Only supported by netcpu_procstat.
3355 during the test as measured by netperf. Only supported by
3360 the test as measured by netperf. Only supported by
3365 the test as measured by netperf. Only supported by
3369 This will display the method used by netperf to measure CPU
3374 unit of work, as measured by netperf. Units: microseconds of CPU
3435 This will display the number of test iterations netperf undertook,
3449 achieved for overall CPU utilization on the system running netperf
3483 data connection during the test as determined by netperf. A value
3508 The throughput as measured by netperf for the successful "send"
3514 The throughput as measured by netperf for the successful "receive"
3532 The CPU to which netperf was bound, if at all, during the test. A
3533 value of -1 means that netperf was not explicitly bound to a CPU
3537 The number of CPUs (cores, threads) detected by netperf. Units:
3542 as measured by netperf. This can be used to see if any one CPU of a
3549 determined by netperf. Units: CPU ID.
3553 running netperf. Assumes all processors in the system (as
3554 perceived by netperf) on which netperf is running are the same
3558 The frequency of the processor(s) on the system running netperf, at
3559 the time netperf made the call. Assumes that all processors
3560 present in the system running netperf are running at the same
3565 value of -1 means that netperf was not explicitly bound to a CPU
3597 netperf was bound. A value of 0 means the data socket was not
3601 The name/address to which the data socket created by netperf was
3606 The address family to which the data socket created by netperf was
3625 The number of successful "send" calls made by netperf against its
3629 The number of successful "receive" calls made by netperf against
3633 The average number of bytes per "receive" call made by netperf
3637 The average number of bytes per "send" call made by netperf against
3641 The number of bytes successfully sent by netperf through its data
3645 The number of bytes successfully received by netperf through its
3649 The sum of bytes sent and received by netperf through its data
3653 The offset from the alignment of the buffers passed by netperf in
3658 The offset from the alignment of the buffers passed by netperf in
3663 The alignment of the buffers passed by netperf in its "send" calls
3668 The alignment of the buffers passed by netperf in its "receive"
3673 The "width" of the ring of buffers through which netperf cycles as
3676 the time the data socket is created. Can be used to make netperf
3680 The "width" of the ring of buffers through which netperf cycles as
3684 make netperf more processor data cache unfriendly. Units: number
3688 The number of bytes to "dirty" (write to) before netperf makes a
3691 prior to building netperf. Units: Bytes.
3694 The number of bytes to "dirty" (write to) before netperf makes a
3697 to building netperf. Units: Bytes.
3700 The number of bytes netperf should read "cleanly" before making a
3703 prior to building netperf. Clean reads start were dirty writes
3708 delay" (eg TCP_NODELAY) option on the data socket used by netperf
3714 by netperf as requested via the test-specific `-C' option. 1 means
3753 netserver rather than netperf.
3757 netperf was running. Units: ASCII Text
3760 The model name of the system on which netperf was running. Units:
3765 netperf was running. Units: ASCII Text
3768 The version number of the OS running on the system on which netperf
3772 The machine architecture of the machine on which netperf was
3785 netserver rather than netperf.
3789 connection went on the system running netperf. Example: eth0.
3794 traffic on the data connection went on the system running netperf.
3800 traffic on the data connection went on the system running netperf.
3806 traffic on the data connection went on the system running netperf.
3812 traffic on the data connection went on the system running netperf.
3819 running netperf. Units: ASCII Text.
3824 system running netperf. Units: ASCII Text.
3829 system running netperf. Units: ASCII Text.
3833 traffic on the data connection went on the system running netperf.
3838 on the data connection went on the system running netperf. Units:
3861 netserver rather than netperf.
3865 transactions) were attempted by netperf. Specified by the global
3867 specified with the configure command prior to building netperf.
3873 on the test) which were attempted by netperf each
3876 with the configure command prior to building netperf. Units:
3883 specified with the configure command prior to building netperf.
3889 on the test) which were attempted by netperf each
3892 with the configure command prior to building netperf. Units:
3923 generated automagically by netperf, or passed to netperf via an
3930 netperf side when the global `-j' option was specified. Units:
3936 netperf side when the global `-j' option was specified. Units:
3942 measured on the netperf side when the global `-j' option was
3948 measured on the netperf side when the global `-j' option was
3954 measured on the netperf side when the global `-j' option was
3960 netperf side when the global `-j' option was specified. Units:
3966 measured on the netperf side when the global `-j' option was
3970 The full command line used when invoking netperf. Units: ASCII
3978 File: netperf.info, Node: Other Netperf Tests, Next: Address Resolution, Prev: The Omni Tests, …
3983 Apart from the typical performance tests, netperf contains some tests
3994 File: netperf.info, Node: CPU rate calibration, Next: UUID Generation, Prev: Other Netperf Tests…
3999 Some of the CPU utilization measurement mechanisms of netperf work by
4002 system is running a netperf test. The ratio of those rates is used to
4005 This means that netperf must know the rate at which the counter
4007 know the rate, netperf will measure it before starting a data transfer
4009 remote systems, and if repeated for each netperf test would make taking
4012 Thus, the netperf CPU utilization options `-c' and and `-C' can take
4015 netperf can be used to run special tests which only report the value of
4019 environment variable and use that in subsequent netperf commands:
4021 LOC_RATE=`netperf -t LOC_CPU`
4022 REM_RATE=`netperf -H <remote> -t REM_CPU`
4023 netperf -H <remote> -c $LOC_RATE -C $REM_RATE ... -- ...
4025 netperf -H <remote> -c $LOC_RATE -C $REM_RATE ... -- ...
4027 If you are going to use netperf to measure aggregate results, it is
4029 values first to avoid issues with some of the aggregate netperf tests
4032 that any one instance of netperf does not know about the other
4033 instances of netperf. It will report global CPU utilization and will
4036 netperf in an aggregate test, but you have to calculate service demands
4040 File: netperf.info, Node: UUID Generation, Prev: CPU rate calibration, Up: Other Netperf Tests
4045 Beginning with version 2.5.0 netperf can generate Universally Unique
4047 $ netperf -t UUID
4054 aggregate netperf test. Say, for instance if they were inserted into a
4058 File: netperf.info, Node: Address Resolution, Next: Enhancing Netperf, Prev: Other Netperf Tests…
4068 hostname from `gethostbyname()' it was generally unlikely and netperf's
4084 test-specific `-H' and `-L' options. At some point, the netperf tests
4086 `getaddrinfo()' - straw-man patches to <netperf-feedback@netperf.org>
4097 File: netperf.info, Node: Enhancing Netperf, Next: Netperf4, Prev: Address Resolution, Up: Top
4103 enhancements to netperf, by all means do so. If you wish to add a new
4104 "suite" of tests to netperf the general idea is to:
4112 3. Edit `src/netperf.c', `netsh.c', and `netserver.c' as required,
4126 <netperf-feedback@netperf.org> or perhaps <netperf-talk@netperf.org>.
4128 <netperf-feedback@netperf.org> is the next step. From that point, it
4133 File: netperf.info, Node: Netperf4, Next: Concept Index, Prev: Enhancing Netperf, Up: Top
4138 Netperf4 is the shorthand name given to version 4.X.X of netperf. This
4139 is really a separate benchmark more than a newer version of netperf,
4140 but it is a descendant of netperf so the netperf name is kept. The
4142 egg-laying-woolly-milk-pig version of netperf :) The more respectful
4143 way to describe it is to say it is the version of netperf with support
4148 on netperf4 are encouraged to join the netperf-dev
4149 (http://www.netperf.org/cgi-bin/mailman/listinfo/netperf-dev) mailing
4151 (http://www.netperf.org/svn/netperf4/trunk).
4154 File: netperf.info, Node: Concept Index, Next: Option Index, Prev: Netperf4, Up: Top
4194 File: netperf.info, Node: Option Index, Prev: Concept Index, Up: Top