Monitoring Results Update

1
Monitoring Results Update
Les Cottrell SLAC Prepared for the ICFA-SCIC, C
ERN Sep 28, 2002

Partially funded by DOE/MICS Field Work Proposal
on Internet End-to-end Performance Monitoring
(IEPM), also supported by IUPAP
2
RTT to world from US

• Note large number of satellite links ( 600ms
  dark red)
  
• Note reduction by Aug 2002

Jan 2000
Aug 2002
3
History - Loss

• Loss more critical than RTT
• Losses cause timeouts of typically seconds

• 40-50 improve/yr
• Best networks below 0.1
• Russia, SE Europe, China several years behind

4
History Loss Quality

• Fewer sites have v. poor to dreadful performance
• More have good performance (

5
History - Throughput quality improvements from US
80 annual improvement factor 10/4yr
TCPBW Factor 100 improvement in 8 years
(1) Macroscopic Behavior of the TCP Congestion
Avoidance Algorithm, Matthis, Semke, Mahdavi,
Ott, Computer Communication Review 27(3), July
1997
6
History one Research ISP
100 growth in traffic/yr for last 12 years
Continuous upgrades Increase packet size (bulk th
roughput apps)
7
History One large Research Site
8
Summary - results

• Internet AR connectivity performance is
  improving
  
• RTT 10-20/yr, loss 50/yr, throughput 80/yr
• Reduced use of satellites, mainly use for new
  hard to get to areas (e.g. S. Russian Republics)
  
  
• China, S.E. Europe, Russia rate of change keeps
  up but several years behind
  
• India, S. America performance is where N. America
  W. Europe were 4 5 years ago
  
• Improvements need constant investments to
  understand improve
  

9
Help

• Looking for better hosts to monitor contacts
  in
  
• Albania, Armenia, Austria, Azerbaijan
• Macedonia, Turkey, Yugoslavia
• Columbia, Venezuela, Cuba, Mexico
• Pakistan
• Africa (apart from Egypt, Uganda South Africa,
  n.b. all 54 countries in Africa now have Internet
  access in capitals)
  
• Note there are a few countries (about 5 of the
  worlds countries) that do not have full Internet
  connections and pay dearly by the byte.
  
• A couple of years ago these included
  Afghanistan, Western Sahara, Christmas Island, S.
  Georgia, Marshall Islands, Myanmar, Montserrat,
  N. Korea, Pitcairn, St Vincente Grenadines

10
ICFA/SCIC Monitoring WG
11
Goals

• Obtain as uniform picture as possible of the
  present performance of the connectivity used by
  the ICFA community
  
• For end of 2002, prepare a report on the
  performance of HEP connectivity, including, where
  possible, the identification of any key
  bottlenecks or problem areas.

12
Administrivia

• ICFA-SCIC-MON web page created
  http//www.slac.stanford.edu/xorg/icfa/scic-netmon
  /
  
• Email list icfa-scic-mon_at_slac.stanford.edu set
  up
  
• Membership

13
Getting started

• Decide on regions (proposal), decide what
  measurements/reports needed for ICFA report
  
• Goals
• A physical region, generally recognized
• Start from UN definitions http//esa.un.org/unpp/d
  efinition.html
  
• Similar connectivity and issues,
• Including countries with HENP requirements, and
  with monitorable connections
  
• Limited number of regions, also sub regions (e.g.
  N. Europe, E. Europe, S. Europe, W. Europe)
  
• Another cut Developed, Less Developed, Least
  Developed
  

14
HENP interest in region

• How important is it to focus efforts on a
  region?
  
• Need estimates to know relevance, for example
• Number of institutes in Particle Physics
  Handbook?
  
• Other suggestions (HENP practitioners/country, do
  we restrict to HENP)?
  
• Then provide contacts/sites/hosts to monitor for
  countries in regions
  
• Some iteration on regions to make them relevant
  to HENP (vs UN).

15
Possible Top Level (10 regions)
Moscow
Siberia
Belorussia
Mongolia
Ukraine
Kazakhstan
Azerbaijan
Georgia
China
Turkey
Iran
Japan
Korea
Israel
Egypt
India
Columbia
Malaysia
Uganda
Peru
Brazil
Australia
Uruguay
S. Africa
Chile
Argentina

• N. America, Latin America (includes Central S.
  America), Europe, S.E. Europe, FSU, Africa, S. E.
  Asia, S. W. Asia, Australasia (incl. Pacific
  Islands).
• Probably need to separate out Japan/Korea/Singapo
  re, Australia/NZ
  

16
Content of Report 1/3

• Methodology overview (refer to earlier report)
• Low impact for developing regions, see previous
  report and PingER pages, deployment
  
• Coverage in time space
• Impact of poor performance
• We also want to address hi-perf throughput (e.g.
  iperf) only relevant for W. Europe, Japan N.
  America?
  

17
Content of Report 2/3

• Results
• Performance of AR nets
• Which ones?
• Hi-perf ESnet, Internet 2, Janet, INFN, DFN,
  CAnet, IN2P3, Renater, Japan
  
• Others
• Performance trends, RTT, loss, derived throughput
  improvements
  
• Current performance tables by regions

18
Losses World by sub-region, Jan 02

• 5bad

• Russia, S America bad
• Balkans, M East, Africa, S Asia, Caucasus poor

19
History - Throughput quality improvements from US
80 annual improvement factor 10/4yr
TCPBW Factor 100 improvement in 8 years
(1) Macroscopic Behavior of the TCP Congestion
Avoidance Algorithm, Matthis, Semke, Mahdavi,
Ott, Computer Communication Review 27(3), July
1997
20
Content of Report 3/3

• Recommendations
• Continued monitoring needed
• Identify what needs extending/improving
• Needs modest funding
• Where is improved performance most needed