QoS on Best-exertion IP Systems.

Uploaded on:
QoS on Best-exertion IP Systems. Les Cottrell – SLAC www.slac.stanford.edu/grp/scs/net/talk/qos-itu-apr01/
Slide 1

QoS on Best-exertion IP Networks Les Cottrell – SLAC www.slac.stanford.edu/grp/scs/net/talk/qos-itu-apr01/Presented at the Joint SG13/SG16 Workshop Panel Session: "Achieving Multimedia QOS over IP-Based Networks" a piece of the ITU-T SG13/SG16 Workshop on IP Networking and Mediacom 2004, Geneva, Switzerland, April 25-27, 2001 Partially financed by DOE/MICS Field Work Proposal on Internet End-to-end Performance Monitoring (IEPM), additionally upheld by IUPAP

Slide 2

Overview Measurement technique Delay Loss Jitter Availability Summary

Slide 3

PingER Measurements from 32 screens in 14 nations Over 600 remote hosts Over 72 nations Over 3300 screen remote website sets Measurements backpedal to Jan-95 Reports on RTT, misfortune, reachability, jitter, reorders, copies … Uses omnipresent “ping” office of TCP/IP Countries checked Contain 78% of world populace 99% of online clients of Internet

Slide 4

RTT from ESnet to Groups of Sites RTT ~ separation/(0.6*c) + jumps * switch delay Router delay = lining + checking in & out + handling ITU G.114 300 ms RTT limit for voice 20%/year

Slide 5

RTT Region to Region OK White 0-64ms Green 64-128ms Yellow 128-256ms NOT OK Pink 256-512ms Red > 512ms OK inside of locales, N. America OK with Europe, Japan

Slide 6

RTT from California to world Europe E. Coast Brazil E. Coast US W. Coast US 300ms RTT (ms) Europe & S. America 3*0.6c Longitude (degrees) 300ms Frequency Source = Palo Alto CA, W. Coast RTT (ms.) Data from CAIDA Skitter venture

Slide 7

Loss seen from US to gatherings of Sites half change/year ETSI DTR/TIPHON-05001 V1.2.5 edge for good discourse

Slide 8

Loss to world from US Using year 2000, division of world’s populace/nation from www.nua.ie/studies/how_many_online/

Slide 9

Losses between Regions

Slide 10

“Jitter” from N. America to W. Europe “Jitter” = IQR(ipdv), where ipdv(i) =RTT(i) – RTT(i-1) 214 sets ETSI: DTR/TIPHON-05001 V1.2.5 (1998-09) great discourse < 75ms jitter

Slide 11

“Jitter” between areas ETSI: DTR/TIPHON-05001 V1.2.5 (1998-09) 125ms=Med 225ms=Poor 75ms=Good Jitter fluctuates with stacking

Slide 12

SLAC-CERN Jitter ETSI/TIPHON delay jitter limit (75 ms)

Slide 13

Availability – Outage Probability Surveyor tests arbitrarily 2/second Measure time ( Outage length ) back to back tests don’t traverse Heavy tailed blackout lengths (parcel misfortune not Poisson) http://www-iepm.slac.stanford.edu/checking/surveyor/outage.html

Slide 14

More Information This discussion: www.slac.stanford.edu/grp/scs/net/talk/qos-itu-apr01/IEPM/PingER home site www-iepm.sla

View more...