1IPERF(1) User Manuals IPERF(1)
2
6 iperf - perform network traffic tests using network sockets. Metrics
7 include throughput and latency.
8
10 iperf -s [options]
11 iperf -c server [options]
13 iperf -u -s [options]
15 iperf -u -c server [options]
17
20 iperf 2 is a testing tool which performs network traffic measurements
21 using network sockets. The performance metrics supported include
22 throughput and latency. Iperf can use both TCP and UDP sockets (or pro‐
23 tocols.) It supports unidirectional, full duplex (same socket) and
24 bidirectional traffic, and supports multiple, simultaneous traffic
25 streams. It supports multicast traffic including source specific multi‐
26 cast (SSM) joins. Its multi-threaded design allows for peak perfor‐
27 mance. Metrics displayed help to characterize host to host network per‐
28 formance. Note: Setting the enhanced (-e) option provides all available
29 metrics.
30 The user must establish both a both a server (to receive traffic) and a
32 client (to generate and send traffic) for a test to occur. The client
33 and server typically are on different hosts or computers but need not
34 be.
35
37 -b, --bandwidth
38 set the target bandwidth and optional standard deviation per
39 <mean>,[<stdev>] (See NOTES for suffixes)
40 -e, --enhanced
42 Display enhanced output in reports otherwise use legacy report
43 (ver 2.0.5) formatting (see NOTES)
44 -f, --format [abkmgBKMG]
46 format to report: adaptive, bits, Bytes, Kbits, Mbits, Gbits,
47 KBytes, MBytes, GBytes (see NOTES for more)
48 -h, --help
50 print a help synopsis
51 --hide-ips
53 obscure ip addresses in output (useful when wanting to publish
54 results and not display the full ip addresses. v4 only)
55 -i, --interval < t | f >
57 sample or display interval reports every t seconds (default) or
58 every frame or burst, i.e. if f is used then the interval will
59 be each frame or burst. The frame interval reporting is experi‐
60 mental. Also suggest a compile with fast-sampling, i.e. ./con‐
61 figure --enable-fastsampling
62 -l, --len n[kmKM]
64 set read/write buffer size (TCP) or length (UDP) to n (TCP de‐
65 fault 128K, UDP default 1470)
66 --l2checks
68 perform layer 2 length checks on received UDP packets (requires
69 systems that support packet sockets, e.g. Linux)
70 -m, --print_mss
72 print TCP maximum segment size (MTU - TCP/IP header)
73 --NUM_REPORT_STRUCTS <count>
75 Override the default shared memory size between the traffic
76 thread(s) and reporter thread in order to mitigate mutex lock
77 contentions. The default value of 5000 should be sufficient for
78 1Gb/s networks. Increase this upon seeing the Warning message of
79 reporter thread too slow. If the Warning message isn't seen,
80 then increasing this won't have any significant effect (other
81 than to use some additional memory.)
82 -o, --output filename
84 output the report or error message to this specified file
85 --permit-key [=<value>]
87 Set a key value that must match for the server to accept traffic
88 on a connection. If the option is given without a value on the
89 server a key value will be autogenerated and displayed in its
90 initial settings report. The lifetime of the key is set using
91 --permit-key-timeout and defaults to twenty seconds. The value
92 is required on clients. The value will also be used as part of
93 the transfer id in reports. The option set on the client but not
94 the server will also cause the server to reject the client's
95 traffic. TCP only, no UDP support.
96 -p, --port m[-n]
98 set client or server port(s) to send or listen on per m (default
99 5001) w/optional port range per m-n (e.g. -p 6002-6008) (see
100 NOTES)
101 --sum-dstip
103 sum traffic threads based upon the destination IP address (de‐
104 fault is source ip address)
105 --sum-only
107 set the output to sum reports only. Useful for -P at large val‐
108 ues
109 -t, --time n
111 time in seconds to listen for new traffic connections, receive
112 traffic or send traffic
113 -u, --udp
115 use UDP rather than TCP
116 -w, --window n[kmKM]
118 TCP window size (socket buffer size)
119 -z, --realtime
121 Request real-time scheduler, if supported.
122 -B, --bind host[:port][%dev]
124 bind to host, ip address or multicast address, optional port or
125 device (see NOTES)
126 -C, --compatibility
128 for use with older versions does not sent extra msgs
129 -M, --mss n
131 set TCP maximum segment size (MTU - 40 bytes)
132 -N, --nodelay
134 set TCP no delay, disabling Nagle's Algorithm
135 -v, --version
137 print version information and quit
138 -x, --reportexclude [CDMSV]
140 exclude C(connection) D(data) M(multicast) S(settings) V(server)
141 reports
142 -y, --reportstyle C|c
144 if set to C or c report results as CSV (comma separated values)
145 -Z, --tcp-congestion
147 Set the default congestion-control algorithm to be used for new
148 connections. Platforms must support setsockopt's TCP_CONGESTION.
149 (Notes: See sysctl and tcp_allowed_congestion_control for avail‐
150 able options. May require root privileges.)
151
153 -1, --singleclient
154 set the server to process only one client at a time
155 -b, --bandwidth n[kmgKMG]
157 set target read rate to n bits/sec. TCP only for the server.
158 -s, --server
160 run in server mode
161 --histograms[=binwidth[u],bincount,[lowerci],[upperci]]
163 enable latency histograms for udp packets (-u), for tcp writes
164 (with --trip-times), or for either udp or tcp with --isochronous
165 clients. The binning can be modified. Bin widths (default 1 mil‐
166 lisecond, append u for microseconds, m for milliseconds) bin‐
167 count is total bins (default 1000), ci is confidence interval
168 between 0-100% (default lower 5%, upper 95%, 3 stdev 99.7%)
169 --permit-key [=<value>]
171 Set a key value that must match for the server to accept traffic
172 from a client (also set with --permit-key.) The server will au‐
173 togenerate a globally unique key when the option is given with‐
174 out a value. This value will be displayed in the server's ini‐
175 tial settings report. The lifetime of the key is set using
176 --permit-key-timeout and defaults to twenty seconds. TCP only,
177 no UDP support.
178 --permit-key-timeout <value>
180 Set the lifetime of the permit key in seconds. Defaults to 20
181 seconds if not set. A value of zero will disable the timer.
182 --tcp-rx-window-clamp n[kmKM]
184 Set the socket option of TCP_WINDOW_CLAMP, units is bytes.
185 -t, --time n
187 time in seconds to listen for new traffic connections and/or re‐
188 ceive traffic (defaults to infinite)
189 -B, --bind ip | ip%device
191 bind src ip addr and optional src device for receiving
192 -D, --daemon
194 run the server as a daemon. On Windows this will run the speci‐
195 fied command-line under the IPerfService, installing the service
196 if necessary. Note the service is not configured to auto-start
197 or restart - if you need a self-starting service you will need
198 to create an init script or use Windows "sc" commands.
199 -H, --ssm-host host
201 Set the source host (ip addr) per SSM multicast, i.e. the S of
202 the S,G
203 -R, --remove
205 remove the IPerfService (Windows only).
206 -U, --single_udp
208 run in single threaded UDP mode
209 -V, --ipv6_domain
211 Enable IPv6 reception by setting the domain and socket to
212 AF_INET6 (Can receive on both IPv4 and IPv6)
213
215 -b, --bandwidth n[kmgKMG][,n[kmgKMG]] | n[kmgKMG]pps
216 set target bandwidth to n bits/sec (default 1 Mbit/sec) or n
217 packets per sec. This may be used with TCP or UDP. Optionally,
218 for variable loads, use format of mean,standard deviation
219 -c, --client host | host%device
221 run in client mode, connecting to host where the optional %dev
222 will SO_BINDTODEVICE that output interface (requires root and
223 see NOTES)
224 --burst-period[=n]
226 Set the burst period in seconds. Defaults to one second. (Note:
227 assumed use case is low duty cycle traffic bursts)
228 --burst-size[=n]
230 Set the burst size in bytes. Defaults to 1M if no value is
231 given.
232 --connect-only[=n]
234 only perform a TCP connect (or 3WHS) without any data transfer,
235 useful to measure TCP connect() times. Optional value of n is
236 the total number of connects to do (zero is run forever.) Note
237 that -i will rate limit the connects where -P will create bursts
238 and -t will end the client and hence end its connect attempts.
239 --connect-retries[= n]
241 number of times to retry a TCP connect at the application level.
242 See operating system information on the details of TCP connect
243 related settings.
244 -d, --dualtest
246 Do a bidirectional test simultaneous test using two unidirec‐
247 tional sockets
248 --fq-rate n[kmgKMG]
250 Set a rate to be used with fair-queueing based socket-level pac‐
251 ing, in bytes or bits per second. Only available on platforms
252 supporting the SO_MAX_PACING_RATE socket option. (Note: Here the
253 suffixes indicate bytes/sec or bits/sec per use of uppercase or
254 lowercase, respectively)
255 --full-duplex
257 run a full duplex test, i.e. traffic in both transmit and re‐
258 ceive directions using the same socket
259 --histograms[=binwidth[u],bincount,[lowerci],[upperci]]
261 enable select()/write() histograms with --tcp-write-prefetch.
262 The binning can be modified. Bin widths (default 100 microsec‐
263 onds, append u for microseconds, m for milliseconds) bincount is
264 total bins (default 10000), ci is confidence interval between
265 0-100% (default lower 5%, upper 95%, 3 stdev 99.7%)
266 --incr-dstip
268 increment the destination ip address when using the parallel
269 (-P) option
270 --incr-dstport
272 increment the destination port when using the parallel (-P) op‐
273 tion
274 --incr-srcip
276 increment the source ip address when using the parallel (-P) op‐
277 tion
278 --ipg n
280 set the inter-packet gap to n (units of seconds) for packets or
281 within a frame/burst when --isochronous is set
282 --isochronous[=fps:mean,stdev]
284 send isochronous traffic with frequency frames per second and
285 load defined by mean and standard deviation using a log normal
286 distribution, defaults to 60:20m,0. (Note: Here the suffixes in‐
287 dicate bytes/sec or bits/sec per use of uppercase or lowercase,
288 respectively. Also the p suffix is supported to set the burst
289 size in packets, e.g. isochronous=2:25p will send two 25 packet
290 bursts every second, or one 25 packet burst every 0.5 seconds.)
291 --local-only[=1|0]
293 Set 1 to limit traffic to the local network only (through the
294 use of SO_DONTROUTE) set to zero otherwise with optional over‐
295 ride of compile time default (see configure --default-localonly)
296 --near-congestion[=n]
298 Enable TCP write rate limiting per the sampled RTT. The delay is
299 applied after the -l number of bytes have completed. The op‐
300 tional value is the multiplier to the RTT and defines the time
301 delay. This value defaults to 0.5 if it is not set. Values less
302 than 1 are supported but the value cannot be negative. This is
303 an experimental feature. It is not likely stable on live net‐
304 works. Suggested use is over controlled test networks.
305 --no-connect-sync
307 By default, parallel traffic threads (per -P greater than 1)
308 will synchronize after their TCP connects and prior to each
309 sending traffic, i.e. all the threads first complete (or error)
310 the TCP 3WHS before any traffic thread will start sending. This
311 option disables that synchronization such that each traffic
312 thread will start sending immediately after completing its suc‐
313 cessful connect.
314 --no-udp-fin
316 Don't perform the UDP final server to client exchange which
317 means there won't be a final server report displayed on the
318 client. All packets per the test will be from the client to the
319 server and no packets should be sent in the other direction.
320 It's highly suggested that -t be set on the server if this op‐
321 tion is being used. This is because there will be only one
322 trigger ending packet sent from client to server and if it's
323 lost then the server will continue to run. (Requires ver 2.0.14
324 or better)
325 -n, --num n[kmKM]
327 number of bytes to transmit (instead of -t)
328 --permit-key [=<value>]
330 Set a key value that must match the server's value (also set
331 with --permit-key) in order for the server to accept traffic
332 from the client. TCP only, no UDP support.
333 -r, --tradeoff
335 Do a bidirectional test individually - client-to-server, fol‐
336 lowed by a reversed test, server-to-client
337 --tcp-write-prefetch n[kmKM]
339 Set TCP_NOTSENT_LOWAT on the socket and use event based writes
340 per select() on the socket.
341 -t, --time n|0
343 time in seconds to transmit traffic, use zero for infinite (de‐
344 fault is 10 secs)
345 --trip-times
347 enable the measurement of end to end write to read latencies
348 (client and server clocks must be synchronized)
349 --txdelay-time
351 time in seconds to hold back or delay after the TCP connect and
352 prior to the socket writes. For UDP it's the delay between the
353 traffic thread starting and the first write.
354 --txstart-time n.n
356 set the txstart-time to n.n using unix or epoch time format
357 (supports microsecond resolution, e.g 1536014418.123456) An ex‐
358 ample to delay one second using command substitution is iperf -c
359 192.168.1.10 --txstart-time $(expr $(date +%s) + 1).$(date +%N)
360 -B, --bind ip | ip:port | ipv6 -V | [ipv6]:port -V
362 bind src ip addr and optional port as the source of traffic (see
363 NOTES)
364 -F, --fileinput name
366 input the data to be transmitted from a file
367 -I, --stdin
369 input the data to be transmitted from stdin
370 -L, --listenport n
372 port to receive bidirectional tests back on
373 -P, --parallel n
375 number of parallel client threads to run
376 -R, --reverse
378 reverse the traffic flow (useful for testing through firewalls,
379 see NOTES)
380 -S, --tos
382 set the socket's IP_TOS (byte) field
383 -T, --ttl n
385 time-to-live, for multicast (default 1) -V, --ipv6_domain Set
386 the domain to IPv6 (send packets over IPv6)
387 -X, --peerdetect
389 run peer version detection prior to traffic.
390 -Z, --linux-congestion algo
392 set TCP congestion control algorithm (Linux only)
393
395 TCP tests (client)
396 iperf -c <host> -e -i 1
398 ------------------------------------------------------------
399 Client connecting to <host>, TCP port 5001 with pid 5149
400 Write buffer size: 128 KByte
401 TCP window size: 340 KByte (default)
402 ------------------------------------------------------------
403 [ 3] local 45.56.85.133 port 49960 connected with 45.33.58.123 port
404 5001 (ct=3.23 ms)
405 [ ID] Interval Transfer Bandwidth Write/Err Rtry
406 Cwnd/RTT NetPwr
407 [ 3] 0.00-1.00 sec 126 MBytes 1.05 Gbits/sec 1006/0 0
408 56K/626 us 210636.47
409 [ 3] 1.00-2.00 sec 138 MBytes 1.15 Gbits/sec 1100/0 299
410 483K/3884 us 37121.32
411 [ 3] 2.00-3.00 sec 137 MBytes 1.15 Gbits/sec 1093/0 24
412 657K/5087 us 28162.31
413 [ 3] 3.00-4.00 sec 126 MBytes 1.06 Gbits/sec 1010/0 284
414 294K/2528 us 52366.58
415 [ 3] 4.00-5.00 sec 117 MBytes 980 Mbits/sec 935/0 373
416 487K/2025 us 60519.66
417 [ 3] 5.00-6.00 sec 144 MBytes 1.20 Gbits/sec 1149/0 2
418 644K/3570 us 42185.36
419 [ 3] 6.00-7.00 sec 126 MBytes 1.06 Gbits/sec 1011/0 112
420 582K/5281 us 25092.56
421 [ 3] 7.00-8.00 sec 110 MBytes 922 Mbits/sec 879/0 56
422 279K/1957 us 58871.89
423 [ 3] 8.00-9.00 sec 127 MBytes 1.06 Gbits/sec 1014/0 46
424 483K/3372 us 39414.89
425 [ 3] 9.00-10.00 sec 132 MBytes 1.11 Gbits/sec 1054/0 0
426 654K/3380 us 40872.75
427 [ 3] 0.00-10.00 sec 1.25 GBytes 1.07 Gbits/sec 10251/0 1196
428 -1K/3170 us 42382.03
429 where (per -e,)
432 ct= TCP connect time (or three way handshake time 3WHS)
433 Write/Err Total number of successful socket writes. Total number
434 of non-fatal socket write errors
435 Rtry Total number of TCP retries
436 Cwnd/RTT (*nix only) TCP congestion window and round trip time
437 (sampled where NA indicates no value)
438 NetPwr (*nix only) Network power defined as (throughput / RTT)
439 TCP tests (server)
442 iperf -s -e -i 1 -l 8K
444 ------------------------------------------------------------
445 Server listening on TCP port 5001 with pid 13430
446 Read buffer size: 8.00 KByte
447 TCP window size: 85.3 KByte (default)
448 ------------------------------------------------------------
449 [ 4] local 45.33.58.123 port 5001 connected with 45.56.85.133 port
450 49960
451 [ ID] Interval Transfer Bandwidth Reads
452 Dist(bin=1.0K)
453 [ 4] 0.00-1.00 sec 124 MBytes 1.04 Gbits/sec 22249
454 798:2637:2061:767:2165:1563:589:11669
455 [ 4] 1.00-2.00 sec 136 MBytes 1.14 Gbits/sec 24780
456 946:3227:2227:790:2427:1888:641:12634
457 [ 4] 2.00-3.00 sec 137 MBytes 1.15 Gbits/sec 24484
458 1047:2686:2218:810:2195:1819:728:12981
459 [ 4] 3.00-4.00 sec 126 MBytes 1.06 Gbits/sec 20812
460 863:1353:1546:614:1712:1298:547:12879
461 [ 4] 4.00-5.00 sec 117 MBytes 984 Mbits/sec 20266
462 769:1886:1828:589:1866:1350:476:11502
463 [ 4] 5.00-6.00 sec 143 MBytes 1.20 Gbits/sec 24603
464 1066:1925:2139:822:2237:1827:744:13843
465 [ 4] 6.00-7.00 sec 126 MBytes 1.06 Gbits/sec 22635
466 834:2464:2249:724:2269:1646:608:11841
467 [ 4] 7.00-8.00 sec 110 MBytes 921 Mbits/sec 21107
468 842:2437:2747:592:2871:1903:496:9219
469 [ 4] 8.00-9.00 sec 126 MBytes 1.06 Gbits/sec 22804
470 1038:1784:2639:656:2738:1927:573:11449
471 [ 4] 9.00-10.00 sec 133 MBytes 1.11 Gbits/sec 23091
472 1088:1654:2105:710:2333:1928:723:12550
473 [ 4] 0.00-10.02 sec 1.25 GBytes 1.07 Gbits/sec 227306
474 9316:22088:21792:7096:22893:17193:6138:120790
475 where (per -e,)
477 Reads Total number of socket reads
478 Dist(bin=size) Eight bin histogram of the socket reads returned
479 byte count. Bin width is set per size. Bins are separated by a
480 colon. In the example, the bins are 0-1K, 1K-2K, .., 7K-8K.
481 TCP tests (server with --trip-times on client) iperf -s -i 1 -w 4M
484 ------------------------------------------------------------
485 Server listening on TCP port 5001
486 TCP window size: 8.00 MByte (WARNING: requested 4.00 MByte)
487 ------------------------------------------------------------
488 [ 4] local 192.168.1.4%eth0 port 5001 connected with 192.168.1.7 port
489 44798 (trip-times) (MSS=1448) (peer 2.0.14-alpha)
490 [ ID] Interval Transfer Bandwidth Burst Latency
491 avg/min/max/stdev (cnt/size) inP NetPwr Reads=Dist
492 [ 4] 0.00-1.00 sec 19.0 MBytes 159 Mbits/sec
493 52.314/10.238/117.155/19.779 ms (151/131717) 1.05 MByte 380.19
494 781=306:253:129:48:18:15:8:4
495 [ 4] 1.00-2.00 sec 20.0 MBytes 168 Mbits/sec
496 53.863/21.264/79.252/12.277 ms (160/131080) 1.08 MByte 389.38
497 771=294:236:126:60:18:24:10:3
498 [ 4] 2.00-3.00 sec 18.2 MBytes 153 Mbits/sec
499 58.718/22.000/137.944/20.397 ms (146/130964) 1.06 MByte 325.64
500 732=299:231:98:52:18:19:10:5
501 [ 4] 3.00-4.00 sec 19.7 MBytes 165 Mbits/sec 50.448/
502 8.921/82.728/14.627 ms (158/130588) 997 KByte 409.00
503 780=300:255:121:58:15:18:7:6
504 [ 4] 4.00-5.00 sec 18.8 MBytes 158 Mbits/sec
505 53.826/11.169/115.316/15.541 ms (150/131420) 1.02 MByte 366.24
506 761=302:226:134:52:22:17:7:1
507 [ 4] 5.00-6.00 sec 19.5 MBytes 164 Mbits/sec
508 50.943/11.922/76.134/14.053 ms (156/131276) 1.03 MByte 402.00
509 759=273:246:149:45:16:18:4:8
510 [ 4] 6.00-7.00 sec 18.5 MBytes 155 Mbits/sec
511 57.643/10.039/127.850/18.950 ms (148/130926) 1.05 MByte 336.16
512 710=262:228:133:37:16:20:8:6
513 [ 4] 7.00-8.00 sec 19.6 MBytes 165 Mbits/sec
514 52.498/12.900/77.045/12.979 ms (157/131003) 1.00 MByte 391.78
515 742=288:200:135:68:16:23:4:8
516 [ 4] 8.00-9.00 sec 18.0 MBytes 151 Mbits/sec 58.370/
517 8.026/150.243/21.445 ms (144/131255) 1.06 MByte 323.81
518 716=268:241:108:51:20:17:8:3
519 [ 4] 9.00-10.00 sec 18.4 MBytes 154 Mbits/sec
520 56.112/12.419/79.790/13.668 ms (147/131194) 1.05 MByte 343.70
521 822=330:303:120:26:16:14:9:4
522 [ 4] 10.00-10.06 sec 1.03 MBytes 146 Mbits/sec
523 69.880/45.175/78.754/10.823 ms (9/119632) 1.74 MByte 260.40
524 62=26:30:5:1:0:0:0:0
525 [ 4] 0.00-10.06 sec 191 MBytes 159 Mbits/sec 54.183/
526 8.026/150.243/16.781 ms (1526/131072) 1.03 MByte 366.98
527 7636=2948:2449:1258:498:175:185:75:48
528 where (per -e,)
530 Burst Latency One way TCP write() to read() latency in mean/min‐
531 imum/maximum/standard deviation format (Note: requires the
532 client's and server's system clocks to be synchronized to a com‐
533 mon reference, e.g. using precision time protocol PTP. A GPS
534 disciplined OCXO is a recommended reference.)
535 cnt Number of completed bursts received and used for the burst
536 latency calculations
537 size Average burst size in bytes (computed average and estimate
538 only)
539 inP inP, short for in progress, is the average number of bytes
540 in progress or in flight. This is taken from the application
541 level write to read perspective. Note this is a mean value. The
542 parenthesis value is the standard deviation from the mean. (Re‐
543 quires --trip-times on client. See Little's law in NOTES.)
544 NetPwr Network power defined as (throughput / one way latency)
545 TCP tests (with one way delay sync check -X and --trip-times on the
548 client)
549 iperf -c 192.168.1.4 -X -e --trip-times -i 1 -t 2
551 ------------------------------------------------------------
552 Client connecting to 192.168.1.4, TCP port 5001 with pid 16762 (1
553 flows)
554 Write buffer size: 131072 Byte
555 TCP window size: 85.0 KByte (default)
556 ------------------------------------------------------------
557 [ 1] Clock sync check (ms): RTT/Half=(3.361/1.680) OWD-
558 send/ack/asym=(2.246/1.115/1.131)
559 [ 1] local 192.168.1.1%ap0 port 47466 connected with 192.168.1.4 port
560 5001 (MSS=1448) (trip-times) (sock=3) (peer 2.1.4-master)
561 [ ID] Interval Transfer Bandwidth Write/Err Rtry
562 Cwnd/RTT NetPwr
563 [ 1] 0.00-1.00 sec 9.50 MBytes 79.7 Mbits/sec 77/0 0
564 2309K/113914 us 87
565 [ 1] 1.00-2.00 sec 7.12 MBytes 59.8 Mbits/sec 57/0 0
566 2492K/126113 us 59
567 [ 1] 2.00-2.42 sec 128 KBytes 2.47 Mbits/sec 2/0 0
568 2492K/126113 us 2
569 [ 1] 0.00-2.42 sec 16.8 MBytes 58.0 Mbits/sec 136/0 0
570 2492K/126113 us 57
571 UDP tests (client)
574 iperf -c <host> -e -i 1 -u -b 10m
576 ------------------------------------------------------------
577 Client connecting to <host>, UDP port 5001 with pid 5169
578 Sending 1470 byte datagrams, IPG target: 1176.00 us (kalman adjust)
579 UDP buffer size: 208 KByte (default)
580 ------------------------------------------------------------
581 [ 3] local 45.56.85.133 port 32943 connected with 45.33.58.123 port
582 5001
583 [ ID] Interval Transfer Bandwidth Write/Err PPS
584 [ 3] 0.00-1.00 sec 1.19 MBytes 10.0 Mbits/sec 852/0 851 pps
585 [ 3] 1.00-2.00 sec 1.19 MBytes 10.0 Mbits/sec 850/0 850 pps
586 [ 3] 2.00-3.00 sec 1.19 MBytes 10.0 Mbits/sec 850/0 850 pps
587 [ 3] 3.00-4.00 sec 1.19 MBytes 10.0 Mbits/sec 851/0 850 pps
588 [ 3] 4.00-5.00 sec 1.19 MBytes 10.0 Mbits/sec 850/0 850 pps
589 [ 3] 5.00-6.00 sec 1.19 MBytes 10.0 Mbits/sec 850/0 850 pps
590 [ 3] 6.00-7.00 sec 1.19 MBytes 10.0 Mbits/sec 851/0 850 pps
591 [ 3] 7.00-8.00 sec 1.19 MBytes 10.0 Mbits/sec 850/0 850 pps
592 [ 3] 8.00-9.00 sec 1.19 MBytes 10.0 Mbits/sec 851/0 850 pps
593 [ 3] 0.00-10.00 sec 11.9 MBytes 10.0 Mbits/sec 8504/0 850 pps
594 [ 3] Sent 8504 datagrams
595 [ 3] Server Report:
596 [ 3] 0.00-10.00 sec 11.9 MBytes 10.0 Mbits/sec 0.047 ms 0/ 8504
597 (0%) 0.537/ 0.392/23.657/ 0.497 ms 850 pps 2329.37
598 where (per -e,)
600 Write/Err Total number of successful socket writes. Total number
601 of non-fatal socket write errors
602 PPS Transmit packet rate in packets per second
603 UDP tests (server) iperf -s -i 1 -w 4M -u
606 ------------------------------------------------------------
607 Server listening on UDP port 5001
608 Receiving 1470 byte datagrams
609 UDP buffer size: 8.00 MByte (WARNING: requested 4.00 MByte)
610 ------------------------------------------------------------
611 [ 3] local 192.168.1.4 port 5001 connected with 192.168.1.1 port 60027
612 (WARN: winsize=8.00 MByte req=4.00 MByte) (trip-times) (0.0) (peer
613 2.0.14-alpha)
614 [ ID] Interval Transfer Bandwidth Jitter Lost/Total
615 Latency avg/min/max/stdev PPS inP NetPwr
616 [ 3] 0.00-1.00 sec 44.5 MBytes 373 Mbits/sec 0.071 ms 52198/83938
617 (62%) 75.185/ 2.367/85.189/14.430 ms 31854 pps 3.64 MByte 620.58
618 [ 3] 1.00-2.00 sec 44.8 MBytes 376 Mbits/sec 0.015 ms
619 59549/143701 (41%) 79.609/75.603/85.757/ 1.454 ms 31954 pps 3.56 MByte
620 590.04
621 [ 3] 2.00-3.00 sec 44.5 MBytes 373 Mbits/sec 0.017 ms
622 59494/202975 (29%) 80.006/75.951/88.198/ 1.638 ms 31733 pps 3.56 MByte
623 583.07
624 [ 3] 3.00-4.00 sec 44.5 MBytes 373 Mbits/sec 0.019 ms
625 59586/262562 (23%) 79.939/75.667/83.857/ 1.145 ms 31767 pps 3.56 MByte
626 583.57
627 [ 3] 4.00-5.00 sec 44.5 MBytes 373 Mbits/sec 0.081 ms
628 59612/322196 (19%) 79.882/75.400/86.618/ 1.666 ms 31755 pps 3.55 MByte
629 584.40
630 [ 3] 5.00-6.00 sec 44.7 MBytes 375 Mbits/sec 0.064 ms
631 59571/381918 (16%) 79.767/75.571/85.339/ 1.556 ms 31879 pps 3.56 MByte
632 588.02
633 [ 3] 6.00-7.00 sec 44.6 MBytes 374 Mbits/sec 0.041 ms
634 58990/440820 (13%) 79.722/75.662/85.938/ 1.087 ms 31820 pps 3.58 MByte
635 586.73
636 [ 3] 7.00-8.00 sec 44.7 MBytes 375 Mbits/sec 0.027 ms
637 59679/500548 (12%) 79.745/75.704/84.731/ 1.094 ms 31869 pps 3.55 MByte
638 587.46
639 [ 3] 8.00-9.00 sec 44.3 MBytes 371 Mbits/sec 0.078 ms
640 59230/559499 (11%) 80.346/75.514/94.293/ 2.858 ms 31590 pps 3.58 MByte
641 577.97
642 [ 3] 9.00-10.00 sec 44.4 MBytes 373 Mbits/sec 0.073 ms
643 58782/618394 (9.5%) 79.125/75.511/93.638/ 1.643 ms 31702 pps 3.55 MByte
644 588.99
645 [ 3] 10.00-10.08 sec 3.53 MBytes 367 Mbits/sec 0.129 ms
646 6026/595236 (1%) 94.967/80.709/99.685/ 3.560 ms 31107 pps 3.58 MByte
647 483.12
648 [ 3] 0.00-10.08 sec 449 MBytes 374 Mbits/sec 0.129 ms
649 592717/913046 (65%) 79.453/ 2.367/99.685/ 5.200 ms 31776 pps (null)
650 587.91
651 where (per -e,)
654 Latency End to end latency in mean/minimum/maximum/standard de‐
655 viation format (Note: requires the client's and server's system
656 clocks to be synchronized to a common reference, e.g. using pre‐
657 cision time protocol PTP. A GPS disciplined OCXO is a recom‐
658 mended reference.)
659 PPS Received packet rate in packets per second
660 inP inP, short for in progress, is the average number of bytes
661 in progress or in flight. This is taken from an application
662 write to read perspective. (Requires --trip-times on client. See
663 Little's law in NOTES.)
664 NetPwr Network power defined as (throughput / latency)
665 Isochronous UDP tests (client)
668 iperf -c 192.168.100.33 -u -e -i 1 --isochronous=60:100m,10m --realtime
670 ------------------------------------------------------------
671 Client connecting to 192.168.100.33, UDP port 5001 with pid 14971
672 UDP isochronous: 60 frames/sec mean= 100 Mbit/s, stddev=10.0 Mbit/s,
673 Period/IPG=16.67/0.005 ms
674 UDP buffer size: 208 KByte (default)
675 ------------------------------------------------------------
676 [ 3] local 192.168.100.76 port 42928 connected with 192.168.100.33
677 port 5001
678 [ ID] Interval Transfer Bandwidth Write/Err PPS
679 frames:tx/missed/slips
680 [ 3] 0.00-1.00 sec 12.0 MBytes 101 Mbits/sec 8615/0 8493 pps
681 62/0/0
682 [ 3] 1.00-2.00 sec 12.0 MBytes 100 Mbits/sec 8556/0 8557 pps
683 60/0/0
684 [ 3] 2.00-3.00 sec 12.0 MBytes 101 Mbits/sec 8586/0 8586 pps
685 60/0/0
686 [ 3] 3.00-4.00 sec 12.1 MBytes 102 Mbits/sec 8687/0 8687 pps
687 60/0/0
688 [ 3] 4.00-5.00 sec 11.8 MBytes 99.2 Mbits/sec 8468/0 8468 pps
689 60/0/0
690 [ 3] 5.00-6.00 sec 11.9 MBytes 99.8 Mbits/sec 8519/0 8520 pps
691 60/0/0
692 [ 3] 6.00-7.00 sec 12.1 MBytes 102 Mbits/sec 8694/0 8694 pps
693 60/0/0
694 [ 3] 7.00-8.00 sec 12.1 MBytes 102 Mbits/sec 8692/0 8692 pps
695 60/0/0
696 [ 3] 8.00-9.00 sec 11.9 MBytes 100 Mbits/sec 8537/0 8537 pps
697 60/0/0
698 [ 3] 9.00-10.00 sec 11.8 MBytes 99.0 Mbits/sec 8450/0 8450 pps
699 60/0/0
700 [ 3] 0.00-10.01 sec 120 MBytes 100 Mbits/sec 85867/0 8574 pps
701 602/0/0
702 [ 3] Sent 85867 datagrams
703 [ 3] Server Report:
704 [ 3] 0.00-9.98 sec 120 MBytes 101 Mbits/sec 0.009 ms 196/85867
705 (0.23%) 0.665/ 0.083/ 1.318/ 0.174 ms 8605 pps 18903.85
706 where (per -e,)
708 frames:tx/missed/slips Total number of isochronous frames or
709 bursts. Total number of frame ids not sent. Total number of
710 frame slips
711 Isochronous UDP tests (server)
714 iperf -s -e -u --udp-histogram=100u,2000 --realtime
716 ------------------------------------------------------------
717 Server listening on UDP port 5001 with pid 5175
718 Receiving 1470 byte datagrams
719 UDP buffer size: 208 KByte (default)
720 ------------------------------------------------------------
721 [ 3] local 192.168.100.33 port 5001 connected with 192.168.100.76 port
722 42928 isoch (peer 2.0.13-alpha)
723 [ ID] Interval Transfer Bandwidth Jitter Lost/Total
724 Latency avg/min/max/stdev PPS NetPwr Frames/Lost
725 [ 3] 0.00-9.98 sec 120 MBytes 101 Mbits/sec 0.010 ms 196/85867
726 (0.23%) 0.665/ 0.083/ 1.318/ 0.284 ms 8585 pps 18903.85 601/1
727 [ 3] 0.00-9.98 sec T8(f)-PDF:
728 bin(w=100us):cnt(85671)=1:2,2:844,3:10034,4:8493,5:8967,6:8733,7:8823,8:9023,9:8901,10:8816,11:7730,12:4563,13:741,14:1
729 (5.00/95.00%=3/12,Outliers=0,obl/obu=0/0)
730 [ 3] 0.00-9.98 sec F8(f)-PDF:
731 bin(w=100us):cnt(598)=15:2,16:1,17:27,18:68,19:125,20:136,21:103,22:83,23:22,24:23,25:5,26:3
732 (5.00/95.00%=17/24,Outliers=0,obl/obu=0/0)
733 where, Frames/lost Total number of frames (or bursts) received. Total
736 number of bursts lost or error-ed
737 T8-PDF(f) Latency histogram for packets
738 F8-PDF(f) Latency histogram for frames
739
743 Note: The environment variable option settings haven't been maintained
744 well. See the source code if these are of interest.
745
747 Numeric options: Some numeric options support format characters per
748 '<value>c' (e.g. 10M) where the c format characters are k,m,g,K,M,G.
749 Lowercase format characters are 10^3 based and uppercase are 2^n based,
750 e.g. 1k = 1000, 1K = 1024, 1m = 1,000,000 and 1M = 1,048,576
751 Rate limiting: The -b option supports read and write rate limiting at
753 the application level. The -b option on the client also supports vari‐
754 able offered loads through the <mean>,<standard deviation> format, e.g.
755 -b 100m,10m. The distribution used is log normal. Similar for the
756 isochronous option. The -b on the server rate limits the reads. Socket
757 based pacing is also supported using the --fq-rate long option. This
758 will work with the --reverse and --full-duplex options as well.
759 Synchronized clocks: The --trip-times option indicates that the
761 client's and server's clocks are synchronized to a common reference.
762 Network Time Protocol (NTP) or Precision Time Protocol (PTP) are com‐
763 monly used for this. The reference clock(s) error and the synchroniza‐
764 tion protocols will affect the accuracy of any end to end latency mea‐
765 surements.
766 Binding is done at the logical level (ip address or layer 3) using the
768 -B option and at the device (or layer 2) level using the percent (%)
769 separator for both the client and the server. On the client, the -B op‐
770 tion affects the bind(2) system call, and will set the source ip ad‐
771 dress and the source port, e.g. iperf -c <host> -B 192.168.100.2:6002.
772 This controls the packet's source values but not routing. These can be
773 confusing in that a route or device lookup may not be that of the de‐
774 vice with the configured source IP. So, for example, if the IP address
775 of eth0 is used for -B and the routing table for the destination IP ad‐
776 dress resolves the output interface to be eth1, then the host will send
777 the packet out device eth1 while using the source IP address of eth0 in
778 the packet. To affect the physical output interface (e.g. dual homed
779 systems) either use -c <host>%<dev> (requires root) which bypasses this
780 host route table lookup, or configure policy routing per each -B source
781 address and set the output interface appropriately in the policy
782 routes. On the server or receive, only packets destined to -B IP ad‐
783 dress will be received. It's also useful for multicast. For example,
784 iperf -s -B 224.0.0.1%eth0 will only accept ip multicast packets with
785 dest ip 224.0.0.1 that are received on the eth0 interface, while iperf
786 -s -B 224.0.0.1 will receive those packets on any interface, Finally,
787 the device specifier is required for v6 link-local, e.g. -c
788 [v6addr]%<dev> -V, to select the output interface.
789 Reverse, full-duplex, dualtest (-d) and tradeoff (-r): The --reverse
791 (-R) and --full-duplex options can be confusing when compared to the
792 older options of --dualtest (-d) and --tradeoff (-r). The newer options
793 of --reverse and --full-duplex only open one socket and read and write
794 to the same socket descriptor, i.e. use the socket in full duplex mode.
795 The older -d and -r open second sockets in the opposite direction and
796 do not use a socket in full duplex mode. Note that full duplex applies
797 to the socket and not to the network devices and that full duplex sock‐
798 ets are supported by the operating systems regardless if an underlying
799 network supports full duplex transmission and reception. It's sug‐
800 gested to use --reverse if you want to test through a NAT firewall (or
801 -R on non-windows systems). This applies role reversal of the test af‐
802 ter opening the full duplex socket. (Note: Firewall piercing may be
803 required to use -d and -r if a NAT gateway is in the path.)
804 Also, the --reverse -b <rate> setting behaves differently for TCP and
806 UDP. For TCP it will rate limit the read side, i.e. the iperf client
807 (role reversed to act as a server) reading from the full duplex socket.
808 This will in turn flow control the reverse traffic per standard TCP
809 congestion control. The --reverse -b <rate> will be applied on transmit
810 (i.e. the server role reversed to act as a client) for UDP since there
811 is no flow control with UDP. There is no option to directly rate limit
812 the writes with TCP testing when using --reverse.
813 TCP Connect times: The TCP connect time (or three way handshake) can be
815 seen on the iperf client when the -e (--enhanced) option is set. Look
816 for the ct=<value> in the connected message, e.g.in '[ 3] local
817 192.168.1.4 port 48736 connected with 192.168.1.1 port 5001 (ct=1.84
818 ms)' shows the 3WHS took 1.84 milliseconds.
819 Port-range Port ranges are supported using the hyphen notation, e.g.
821 6001-6009. This will cause multiple threads, one per port, on either
822 the listener/server or the client. The user needs to take care that the
823 ports in the port range are available and not already in use per the
824 operating system. The -P is supported on the client and will apply to
825 each destination port within the port range. Finally, this can be used
826 for a workaround for Windows UDP and -P > 1 as Windows doesn't dispatch
827 UDP per a server's connect and the quintuple.
828 Packet per second (pps) calculation The packets per second calculation
830 is done as a derivative, i.e. number of packets divided by time. The
831 time is taken from the previous last packet to the current last packet.
832 It is not the sample interval time. The last packet can land at differ‐
833 ent times within an interval. This means that pps does not have to
834 match rx bytes divided by the sample interval. Also, with --trip-times
835 set, the packet time on receive is set by the sender's write time so
836 pps indicates the end to end pps with --trip-times. The RX pps calcula‐
837 tion is receive side only when -e is set and --trip-times is not set.
838 Little's Law in queuing theory is a theorem that determines the average
840 number of items (L) in a stationary queuing system based on the average
841 waiting time (W) of an item within a system and the average number of
842 items arriving at the system per unit of time (lambda). Mathematically,
843 it's L = lambda * W. As used here, the units are bytes. The arrival
844 rate is taken from the writes.
845 Network power: The network power (NetPwr) metric is experimental. It's
847 a convenience function defined as throughput/delay. For TCP transmits,
848 the delay is the sampled RTT times. For TCP receives, the delay is the
849 write to read latency. For UDP the delay is the end/end latency.
850 Don't confuse this with the physics definition of power (delta en‐
851 ergy/delta time) but more of a measure of a desirable property divided
852 by an undesirable property. Also note, one must use -i interval with
853 TCP to get this as that's what sets the RTT sampling rate. The metric
854 is scaled to assist with human readability.
855 Multicast: Iperf 2 supports multicast with a couple of caveats. First,
857 multicast streams cannot take advantage of the -P option. The server
858 will serialize multicast streams. Also, it's highly encouraged to use a
859 -t on a server that will be used for multicast clients. That is because
860 the single end of traffic packet sent from client to server may get
861 lost and there are no redundant end of traffic packets. Setting -t on
862 the server will kill the server thread in the event this packet is in‐
863 deed lost.
864 Fast Sampling: Use ./configure --enable-fastsampling and then compile
866 from source to enable four digit (e.g. 1.0000) precision in reports'
867 timestamps. Useful for sub-millisecond sampling.
868
870 Use ./configure --enable-thread-debug and then compile from source to
871 enable both asserts and advanced debugging of the tool itself.
872
874 See https://sourceforge.net/p/iperf2/tickets/
875
877 Iperf2, based from iperf (originally written by Mark Gates and Alex
878 Warshavsky), has a goal of maintenance with some feature enhancement.
879 Other contributions from Ajay Tirumala, Jim Ferguson, Jon Dugan <jdugan
880 at x1024 dot net>, Feng Qin, Kevin Gibbs, John Estabrook <jestabro at
881 ncsa.uiuc.edu>, Andrew Gallatin <gallatin at gmail.com>, Stephen Hem‐
882 minger <shemminger at linux-foundation.org>, Tim Auckland <tim.auckland
883 at gmail.com>, Robert J. McMahon <rjmcmahon at rjmcmahon.com>
884
886 accept(2),bind(2),close(2),connect(2),fcntl(2),getpeername(2),getsock‐
887 name(2),getsockopt(2),listen(2),read(2),recv(2),select(2),send(2),set‐
888 sockopt(2),shutdown(2),write(2),ip(7),socket(7),tcp(7),udp(7)
889 Source code at http://sourceforge.net/projects/iperf2/
891 "Unix Network Programming, Volume 1: The Sockets Networking API (3rd
893 Edition) 3rd Edition" by W. Richard Stevens (Author), Bill Fenner (Au‐
894 thor), Andrew M. Rudoff (Author)
895NLANR/DAST July 2021 IPERF(1)