IEPM FY2003 Budget Submission Program Narrative

 

Task Description

 

This task provides for maintenance, enhancement and development of the Internet End-to-end Performance Monitoring (IEPM) tools, measurements, data collection and archiving, analysis and associated reports. In January 2001, the IEPM PingER data gathering tools were deployed at 32 sites in 14 countries. A total of 602 remote nodes at about 420 sites in 72 countries are probed by these monitoring sites, providing performance metrics on over 3300 end-to-end pairs. The 72 countries have over 78% of the world's population and about 99% of the online users of the Internet. PingER is thought to be the largest network performance monitoring project in existence. As with so many aspects of modern computing and networking, the needs of the High Energy Physics (HEP) and ESnet communities continue to push the envelope, and the PingER tools are an important part in measuring the success.

 

 

Progress in FY2000

 

In FY2000, the IEPM project extended the monitoring from 28 to 32 monitoring sites, 2100 to 3300 end-to-end pairs. It also extended the number of Beacon sites (selected subset of all the sites, that are monitored by all measurement sites) from 56 to 74. New visualization tools were introduced including simple Java plots of the tabular data and maps with animated bar charts of performance as a function of time.

 

To assist in scaling up to managing larger numbers of hosts and more complex tool configurations, tools were designed and developed to assist in the management of the configuration databases, i.e., manage the details of the large number of nodes involved in the PingER project, and provide smooth scaling to handle thousands of pairs of sites. Tools were developed to assist in verifying that measurements have been made, the data collected and successfully archived.  The tabular and graphical reports were also enhanced so they scale in the sense they can be generated in an appropriate amount of time, they are of a suitable length, and summaries and subsets are generated that can be appropriately navigated.

 

In co-operation with HEPNRC, SLAC, the timeping program was extended to start to add the concept of directives where the host monitoring site can have greater control over the parameters by specifying the scheduling mechanism and frequency at which to send probes, the size of the probes, the type of probes to send, and the way the results are logged. The ability to customize the frequency and size of the probes will allow finer detailed monitoring of high performance links (which is becoming increasingly important for high performance applications such as those proposed by various Grid teams), and lower impact on poorly provisioned links (such as those in the developing world).

 

Extensive validation of the ping results was performed by comparing and contrasting with results from other complementary measurement projects including AMP, NIMI, Surveyor, and RIPE. To assist in this, probes were hosted from each of these projects. Ways were designed and evaluated to identify and to track the impact of ICMP rate limiting (where ping probes are limited for security or bandwidth reasons and may mistakenly indicate poorer network performance than is the case) and actively worked with such sites or avoid monitoring them.

 

SLAC took a leading role in monitoring performance for the 6bone/6REN IPv6 network. In FY2000, the IPv6 monitoring site at SLAC was put into production and results were presented at the Adelaide IETF.

 

Fifteen presentations on PingER were given at the BaBar collaboration, ESCC and ESSC meetings, the Passive and Active Measurements conference in New Zealand, the IETF in Adelaide, Internet meetings, CAIDA, XIWT/IPERF, and CHEP 2000. Four publications were published in CHEP 2000, the IEEE Communications Magazine on Network Traffic Measurements and Experiments, and in the proceedings of the Internet2 International meeting.

 

 

Expected Progress in FY2001

 

Reports will be added to provide information on duplicate pings and out-of-order pings. Traceroute measurements (traceping), archiving, mining and reporting will be ported from the existing VMS platform to Unix, and the new tools made available publicly, since many sites are reducing their support for VMS. As before, all raw and processed data and results will be made public via the web to enable researchers and networkers to perform their own analyses and accelerate progress and understanding in this important field.

In collaboration with Rice University, SLAC will install, configure, understand, extend and start to use the UCB ns-2 network simulator to further understand Internet performance and dynamics, and allow comparison and contrast with real measurements of throughput, re-ordering etc.

SLAC will assist with installing the SLAC-developed IPv6 monitoring tools at the 6-TAP in Chicago and at the CERNet NOC in China.

The PingER project provided reports and transparencies to the ESnet review at LBNL, the ESnet Annual report, and the International Union of Pure and Applied Physics (to assist in recommending improved connectivity for physicists in developing countries).

SLAC will work with CAIDA to evaluate the effectiveness of hosting a skitter probe at SLAC, and if successful will understand the skitter reports and compare/contrast with PingER.

Early field test applications from the NetPredict start-up will be evaluated (funded under an SBIR) to provide feedback on and suggestions on improving their effectiveness.

Presentations will be made at:

·         CHEP01. The Computing in High Energy Physics Conference to be held in September 2001 is the primary showcase for this work in the HEP community. Two talks have been submitted to present at the meeting, one on Ipv6, the other on high performance throughput.

 

·         Internet2 International. SLAC will detail findings on performance to sites on networks involved in the Internet2 International group. In some cases this will provide a before-and-after picture of the effect of peering.

 

·         PAM2001. An abstract has been submitted and accepted to present a paper at the Passive and Active Monitoring conference in April. The paper will look at how to provide high throughput performance on the Internet.

 

·         ITU. A presentation will be made and a SLAC representative will sit on a panel on Voice over IP (VoIP) and QoS at the ITU meeting in Geneva in April 2001.

 

·         SC2001. SLAC will participate in the Super Computing 2001 in the Fall ofNovember 20010.

 

·         A series of lectures will be presented on “High Performance networking and network measurements” at the Islamabad summer school in June 2001.

 

·SIGCOMM 2001.  SLAC will submit a paper to be titled “Accurate Network Performance Monitoring in the Presence of Differentiated Services.”  This paper will be based on the work to detect signatures of rate limiting and networks using Expedited Forwarding.

 

·         Ipv6 Forum and IETF.  SLAC is involved in performance monitoring for the forthcoming Ipv6 forum summit.  A paper reviewing SLAC’s results will be presented to the Ipv6 Working Group at the next IETF meeting.

 

Two papers were will be submitted for publication, one at the CHEP01, the second at PAM 2001.

 

 

Expected Progress in FY2002 and FY2003

 

The traceroute server tools will be supported and extended.  SLAC will assist in deploying them at critical sites and provide assistance to those using the tools to measure and understand Internet topology, etc. SLAC will provide documentation on the raw data available and on demand compress it and make it available to researchers.  SLAC will continue to run the PingER project, coordinating with the monitoring sites, developing and deploying new and improved measurement and analysis/reporting/display tools. New measurement techniques will be added as options to PingER, i.e., measuring throughput, web access speeds, etc. We will also, and build onto the existing PingER archiving/analysis/ reporting infrastructure. SLAC will provide reports and graphics on Internet performance to various ESnet related activities (ESCC, ESSC, ESnet review committees), and to HENP and other scientific research activities.

 

The ns-2 simulator will be extended and used to try and understand how better to estimate bottle-neck bandwidths and the effects of Internet cross-traffic on high throughput applications.

Troubleshooting diagnostic tools will be improved (such as ping and pingroute), the new Unix traceroute measurement, collection, archiving, data mining and reporting tools will be publicized and distributed, and a simple Java tool will be developed for collecting and graphing ping responses to a small number of hosts in real time.

 

A web site will be designed that contains a collection of illustrative troubles, the attempts to diagnose and, where appropriate, how the troubles y were solved. As this collection grows, a taxonomy of troubles will be developed and navigation tools will be added to assist in matching new troubles to the existing reports.  Documentation will be developed that provides guidance on how to troubleshoot problems from an end-user perspective (where the end user in this case is initially a networker at an end-site).

 

Investigation will continue on how to extend the IPv6 monitoring, analyze the data gathered in terms of trends, and compare and contrast with the exiting Internet (IPv4) performance.

 

Coordinating efforts will be continued and extended with the XIWT, ESnet/ESSC/ESC, Internet 2/Abilene, the IETF/IPPM and HENP.  Assistance will continue to be provided to the ICFA/SCIC group, the IUPAP, and other physics groups as requested.