Search this site with

End User Tools and Projects
- NextINet
- Advanced Applications
Database
- DAST Projects/Tools
- Network Performance
and Measurement Tools

End User Support
- Getting Started Guide
- Networking Glossary
- Other Projects/Organizations
- Funding Opportunities

Documents
- Guides/Tutorials
- Papers/Articles
- Presentations
- Reference Books

WebCT Courses
- Tuning Applications

Events
- NLANR/DAST Training
- NLANR Packets Calendar
- Idesk Travel Schedule

News
- Press Releases
- Alliance Data Link
- I2 Newswire Archives

Reports & Statistics
- Monthly Updates and QSRs
- Abilene "Weather Map"
- Web Server Stats

NLANR Measurement & Operations Analysis Team (MOAT)'s Active Measure Program does site-to-site measurements across the vBNS and complements the vBNS backbone measurements provided by MCI. Currently the measurement includes round-trip-time (latency), packet loss rate, and topology (routing). An on-demand site-to-site throughput test is available via a web page but access is restricted due to possible abuse.

AMP deploys one measurement machine at each participating site. As of July 1999, there are 67 sites. The AMP machines are scheduled to issue ping and traceroute commands to the other machines at regular intervals. The measurement data are shipped to a central machine where post-processing takes place. The results are then available for viewing via the web interface. Normally, traffic to the AMP machines is expected to be in the order of 10-20 MB per day, but may be higher if event-initiated throughput tests are taking place.

After selecting a site on the AMP web page, a list of latencies and losses relative to the selected site appears. Detailed graphs can be obtained by selecting individual sites, which allows for further zooming to individual dates as well as looking into the routing information. A large spike in the latencies and losses usually indicates severe network congestion or problematic routes. Changing of routes over time usually indicates a network topology problem.

An example of round-trip-time measurement from NCSA to NCAR on July 6, 1999, is shown in Figure 1 below. In this figure, the round-trip-time was a relatively constant 30 milliseconds, except during 9:00 am to noon when the round-trip-time was a longer 35 milliseconds, indicating a modest congestion condition. The small spikes in the figure indicate longer delays, which may be the results of running high-throughput applications over the network. Overall, the round-trip-time from NCSA to NCAR on that day was reasonable and didn't expose any problematic situations. For detail information about the AMP results, please read the online user's guide.

Figure 1. AMP round-trip-time measurement from NCSA to NCAR

NLANR and NSF encourage HPC sites to join the AMP program. To request an AMP measurement machine being installed on your campus, submit the web form, and the NLANR measurement team will get in touch with you.