Pioneer: Currently Deployed Tools

Go back to the Pioneer Overview section.

Overview
Currently, the framework allows for measurements between any two Cakeboxes. A limited set of tools (Iperf and traceroute, at this time) are also available for measurements between a small set of Abilene nodes. The following four tools are available in the deployed version of Pioneer, and the following sections describe the constraints placed upon them by our framework to try to limit abuse of the system. We are in the process of adding authentication to Pioneer, which will allow better management of the constraints set. The last section in this document provides a glossary of terms described in the information subsections.

Iperf
traceroute
pchar
ping
Surveyor
Glossary

Iperf Based Output
Originally: The Iperf program was written at NLANR to be a replacement for older tools such as ttcp and nettest.

Information Provided: Latency, loss, jitter. (See Glossary)

Constraints: We currently run Iperf on Abilene and Cakebox nodes for no more than 20 seconds at a time, so as to not affect other users; this is an active measurement tool and perturbs the network. We also do not run multiple Iperf servers on the same machine. If one is already running, our tool returns with an error. Iperf, in this case, is UDP based.

Administrator Options: There is a special unconstrained version of Iperf available to responsible admin type individuals. This version of Iperf also has multicast support. Contact the Pioneer team for more information on how to access this option AFTER you have a fully installed Cakebox at your site.

Output Information: Sample output from this program gives an idea of what is measured. In the last column, the number within parenthesis is the percentage of loss of data sent in that transfer. Lost and Total refer to the number of packets lost, and sent.

-----------------------------------------------------------
Server listening on UDP port 5001
Receiving 1470 byte datagrams
UDP buffer size: 60.0 KByte (default)
------------------------------------------------------------

[ 4] local 172.31.178.168 port 5001 connected with 172.16.7.4 port 9726

[ ID] Interval Transfer Bandwidth Jitter Lost/Total Datagrams

[ 4] 0.0- 1.0 sec 1.3 MBytes 10.0 Mbits/sec 0.209 ms 1/ 894 (0.11%)
[ 4] 1.0- 2.0 sec 1.3 MBytes 10.0 Mbits/sec 0.221 ms 0/ 892 (0%)
[ 4] 2.0- 3.0 sec 1.3 MBytes 10.0 Mbits/sec 0.277 ms 0/ 892 (0%)
[ 4] 3.0- 4.0 sec 1.3 MBytes 10.0 Mbits/sec 0.359 ms 0/ 893 (0%)
[ 4] 4.0- 5.0 sec 1.3 MBytes 10.0 Mbits/sec 0.251 ms 0/ 892 (0%)
[ 4] 5.0- 6.0 sec 1.3 MBytes 10.0 Mbits/sec 0.215 ms 0/ 892 (0%)
[ 4] 6.0- 7.0 sec 1.3 MBytes 10.0 Mbits/sec 0.325 ms 0/ 892 (0%)
[ 4] 7.0- 8.0 sec 1.3 MBytes 10.0 Mbits/sec 0.254 ms 0/ 892 (0%)
[ 4] 8.0- 9.0 sec 1.3 MBytes 10.0 Mbits/sec 0.282 ms 0/ 892 (0%)
[ 4] 0.0-10.0 sec 12.5 MBytes 10.0 Mbits/sec 0.243 ms 1/ 8922 (0.011%)

traceroute Based Output
Originally: Unix networking implementations written by Van Jacobson. Available with all standard installations of Unix based systems.

Information Provided: RTT, route taken between end points. (See Glossary)

Constraints: None. The version run is basic and without additional flags.

Administrator Options: None

Output Information: Three probes are sent to each hop along a path, and the RTT is printed for each one (3 per line). If there is no response within 5 seconds, the output prints a "*". The default maximum number of hops is 30. "!H" means the host cannot be reached.

[yak 71]% traceroute nis.nsf.net.

traceroute to nis.nsf.net (35.1.1.48), 30 hops max, 38 byte packet

1 helios.ee.lbl.gov (128.3.112.1) 19 ms 19 ms 0 ms
2 lilac-dmc.Berkeley.EDU (128.32.216.1) 39 ms 39 ms 19 ms
3 lilac-dmc.Berkeley.EDU (128.32.216.1) 39 ms 39 ms 19 ms
4 ccngw-ner-cc.Berkeley.EDU (128.32.136.23) 39 ms 40 ms 39 ms
5 ccn-nerif22.Berkeley.EDU (128.32.168.22) 39 ms 39 ms 39 ms
6 128.32.197.4 (128.32.197.4) 40 ms 59 ms 59 ms
7 131.119.2.5 (131.119.2.5) 59 ms 59 ms 59 ms
8 129.140.70.13 (129.140.70.13) 99 ms 99 ms 80 ms
9 129.140.71.6 (129.140.71.6) 139 ms 239 ms 319 ms
10 129.140.81.7 (129.140.81.7) 220 ms 199 ms 199 ms
11 nic.merit.edu (35.1.1.48) 239 ms 239 ms 239 ms

pchar Based Output
Originally: Developed by Bruce A. Mah. Based on the algorithms presented as a part of pathchar, which was written by Van Jacobson. Details on pchar availability and use are available here. Information on pathchar can be found here.

Information Provided: Bandwidth, RTT, Queue delay. (See Glossary)

Constraints: Three versions of this tool are available to the user. The one with the highest accuracy can run for as long as an hour. It sends 46 tests per repetition, and 32 repetitions per hop. The mid level test sends 23 tests per repetition, and 24 repetitions per hop. The least accurate and least time consuming test sends 15 tests per repetition, and 20 repetitions per hop. This final low accuracy test will take 5-10 minutes. The other parameters are assumed to be default values.

Administrator Options: None

Output Information: The developer states, "pchar measures network throughput and round-trip time by sending varying-sized UDP packets into the network and waiting for ICMP messages in response." The output here is presented as information per hop. Each hop consists of four lines of output: Partial loss documents the number and percentage of probe packets that were lost during the probes for that hop. The partial char line shows the estimated round-trip time from the probing host through the current hop. The hop char line shows estimates of the round-trip time and bandwidth for the current hop. Finally, the partial queueing shows an estimate of the average queueing along the path, up to and including the current hop.

1: 146.246.243.254 (con243.ca.sandia.gov)

Partial loss:     0 / 1472 (0%)
Partial char:     rtt = 0.811278 ms, (b = 0.000454 ms/B), r2 = 0.995401
                  stddev rtt = 0.003499, stddev b = 0.000005
Hop char:         rtt = 0.154043 ms, bw = 83454.764777 Kbps
Partial queueing: avg = 0.000153 ms (336 bytes)

ping Based Output
Originally: 4.3 BSD Systems

Information Provided: Loss, RTT (See Glossary)

Constraints: We currently run ping on Abilene and Cakebox nodes for no more than 20 packets at a time and for no more than a maximum of 60 seconds, so as to not affect other users. Like Iperf, this is an active measurement tool and perturbs the network. The tool runs in quiet mode, and intermediate results are not printed. This tool will probably be removed once the Surveyor data is available again.

Administrator Options: None

Output Information: The output provides loss as a percentage of packets sent. It also displays the minimum, maximum and average RTT.

--- some-machine.internet2.edu ping statistics ---
20 packets transmitted, 20 packets received, 0% packet loss
round-trip min/avg/max/mdev = 0.203/0.235/0.445/0.054 ms

Surveyor Based Output (Currently unavailable.)
The Surveyor project aims at developing a measurement infrastructure that is deployed at over 100 sites around the world. Currently the primary information available from their sites provides latency and loss information between any two Surveyor nodes. Raw data is collected continuously at these nodes, and a one minute summary is available in real time. Based on this summary, the output from a Pioneer query consists of a table of 5 columns, each of which are described here. The default amount of time for which the data is collected is 30 minutes (going backwards from the time the query was made). The interface lets you pick the path for which results are requested. This path is unidirectional from Source to Dest , and the information is not in any way indicative of the reverse path characteristics of Dest to Source. A "-" value in the output indicates that data is being collected, but not available for the time requested. Our tool returns an error if no data is being collected because of policy or error in the Surveyor programs for a given path.

Date: Summary for a given minute.

Minima: Minimum value of latency in microseconds.

50th Percentile: 50th percentile value of latency in microseconds.

90th Percentile: 90th percentile value of latency in microseconds.

Loss Percentage: Percentage of loss that occured over the 1 minute period for which the summary was created.

Glossary
loss: Number of packets dropped as a percentage of total number of packets sent.

delay:Time taken from when the packet was sent to when it was received.

jitter: Variation in packet delays. Particularly useful for streaming applications such as audio and video, where low jitter is good.

average queue: time spent in queues on path upto and including current hop.

RTT: Round trip time

bandwidth: Amount of data transfered per unit time.