Pathchar measurements from SLAC to top 5 sites

Pathchar

To try and characterize the paths to the top sites of interest to SLAC for high performance bulk thruput, we used pathchar. Pathchar allows a user to find the bandwidth, delay, average queue and loss rate of every hop between a source & destination on the Internet. Pathchar is a useful tool, however, it "does not give exact results, the error is not neglectable, all results between 20 and 30 Mbps have to be considered as equivalent" Gille Farrache - IN2P3

. In all cases shown below the host at SLAC was a Sun running Solaris 2.6 with a 100Mbps Ethernet connection. The information reported for SLAC has been edited to remove information on the internal LAN configuration.

IN2P3 - Lyon, France

The measurements from SLAC to IN2P3 indicate that there are comparable bottlenecks between ESNET-A-GATEWAY at SLAC and the ESnet router at Chicago (26Mb/s), between hop 5 and 6 (24Mb/s) and hops 7 and 8 (21Mb/s). According to Gilles Farrache of IN2P3, the latter two routera are located at CERN, each connected to the same LS1010 with a 155 Mbps fiber, and having a 34 Mbps VC defined between them.

pathchar to ccobsn04.in2p3.fr (134.158.104.144)
 mtu limitted to 1500 bytes at FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 doing 32 probes at each of 64 to 1500 by 44
 0 host.SLAC.Stanford.EDU 
 |    25 Mb/s,   299 us (1.08 ms)
 1 router1.SLAC.Stanford.EDU 
 |    76 Mb/s,   170 us (1.58 ms)
 2 router2.SLAC.Stanford.EDU 
 |   133 Mb/s,   75 us (1.82 ms)
 3 router3.SLAC.Stanford.EDU 
 |   118 Mb/s,   -112 us (1.70 ms)
 4 ESNET-A-GATEWAY.SLAC.Stanford.EDU (192.68.191.18)
                        -> 192.68.191.18 (2)           
 |    26 Mb/s,   28.0 ms (58.2 ms)
 5?chicago1-atms.es.net (134.55.24.17)
 |    24 Mb/s,   575 us (59.9 ms)
 6 206.220.243.32 (206.220.243.32)
                        -> 206.220.243.32 (4)           
 |    43 Mb/s,   54.4 ms (169 ms)
 7?cernh9-s5-0.cern.ch (192.65.184.142)
                        -> 192.65.184.142 (2)           
 |    21 Mb/s,   530 us (171 ms)
 8?in2p3-cernh9.cern.ch (192.65.184.46)
                        -> 192.65.184.46 (1)           
 |    31 Mb/s,   1.48 ms (174 ms)
 9?192.70.69.29 (192.70.69.29)
                        -> 192.70.69.29 (3)           
 |   1838 Mb/s,   37 us (174 ms)
10?Lyon-ANDA.in2p3.fr (134.158.240.1)
                        -> 134.158.240.1 (1)           
 |    48 Mb/s,   -184 us (174 ms)
11?ccobsn04.in2p3.fr (134.158.104.144)
11 hops, rtt 171 ms (174 ms), bottleneck  21 Mb/s, pipe 450913 bytes

The pathchar measurements in the reverse direction, IN2P3 to SLAC, are shown below, and indicate similar bottlenecks. Note that the link between Cern1.in2p3.fr and 192.65.184.150 is a 100Mbps FDDI ring which pathchar asserts is able to run at 169 Mbps!

ccobsn04_root(0)# ./pathchar datamove3.slac.stanford.edu
pathchar to datamove3.slac.stanford.edu (134.79.125.53)
 doing 32 probes at each of 64 to 1500 by 32
 0 localhost
 |    46 Mb/s,   309 us (881 us)
 1 Lyon-ANDA.in2p3.fr (134.158.104.100)
 |    76 Mb/s,   78 us (1.19 ms)
 2 Lyon-INTER.in2p3.fr (134.158.240.4)
 |    25 Mb/s,   1.56 ms (4.79 ms)
 3 Cern1.in2p3.fr (192.70.69.30)
 |   169 Mb/s,   -162 us (4.53 ms),  +q 2.76 ms (58.2 KB) *6
 4 192.65.184.150 (192.65.184.150)
 |    44 Mb/s,   54.9 ms (115 ms),  +q 1.39 ms (7.65 KB) *3
 5 ar1-chicago.cern.ch (192.65.184.141)
 |    32 Mb/s,   666 us (116 ms),  +q 2.61 ms (10.3 KB) *6
 6 chicago-nap.es.net (206.220.243.85)
 |    22 Mb/s,   27.9 ms (173 ms),  +q 6.63 ms (18.6 KB) *5
 7 slac1-atms.es.net (134.55.24.13)
 |   ?? b/s,   336 us (173 ms)
 8 RTR-DMZ.SLAC.Stanford.EDU (192.68.191.17)

Caltech, Pasadena, California, USA

The pathchar measurements from SLAC to Caltech over ESnet appear to show possible bottlenecks at Caltech (20Mbps), between the ESNET-A-GATEWAT at SLAC and the ESnet router at General Atomics (GAC) in San Diego (26Mb/s).

pathchar to heppcs1.cithep.caltech.edu (131.215.126.87)
 mtu limitted to 1500 bytes at FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 doing 32 probes at each of 64 to 1500 by 44
 0 host.SLAC.Stanford.EDU 
 |    25 Mb/s,   295 us (1.07 ms)
 1 router1.SLAC.Stanford.EDU
 |    90 Mb/s,   180 us (1.56 ms)
 2 router2.SLAC.Stanford.EDU
 |    100 Mb/s,   57 us (1.79 ms)
 3 router3.SLAC.Stanford.EDU 
 |   110 Mb/s,   -108 us (1.69 ms)
 4 ESNET-A-GATEWAY.SLAC.Stanford.EDU (192.68.191.18)
 |    26 Mb/s,   8.17 ms (18.5 ms)
 5 gac2-atms.es.net (134.55.24.6)
 |    41 Mb/s,   76 us (18.9 ms)
 6 UCSD--esnet.ATM.calren2.net (198.32.248.69)
 |   106 Mb/s,   1.72 ms (22.5 ms)
 7 USC--UCSD.POS.calren2.net (198.32.248.33)
 |   427 Mb/s,   435 us (23.4 ms)
 8 QAnh--USC.POS.calren2.net (198.32.248.17)
 |   128 Mb/s,   332 us (24.2 ms)
 9 UCI--QAnh.POS.calren2.net (198.32.248.126)
 |   1123 Mb/s,   611 us (25.4 ms)
10 UCR--UCI.POS.calren2.net (198.32.248.13)
 |   ?? b/s,   188 us (25.7 ms)
11 CIT--UCR.POS.calren2.net (198.32.248.9)
 |    73 Mb/s,   619 us (27.1 ms)
12 media-converter.caltech.edu (192.41.208.50)
 |    60 Mb/s,   418 us (28.2 ms)
13 SFL-border.dmz.caltech.edu (192.12.19.252)
 |    20 Mb/s,   1 us (28.8 ms)
14 Lauritsen-RSM.ilan.caltech.edu (131.215.254.112)
 |    36 Mb/s,   -740 us (27.6 ms),  14% dropped
15 heppcs1.cithep.caltech.edu (131.215.126.87)
15 hops, rtt 24.5 ms (27.6 ms), bottleneck  20 Mb/s, pipe 70090 bytes

The pathchar measured on 9/25/00 from SLAC to Caltech is shown below:

 mtu limitted to 1500 bytes at FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 doing 32 probes at each of 64 to 1500 by 44
 0 FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 |    25 Mb/s,   262 us (1.00 ms)
 1 RTR-CORE1.SLAC.Stanford.EDU (134.79.19.2)
 |    80 Mb/s,   173 us (1.50 ms)
 2 RTR-CGB6.SLAC.Stanford.EDU (134.79.135.6)
 |   121 Mb/s,   71 us (1.74 ms)
 3 RTR-DMZ.SLAC.Stanford.EDU (134.79.111.4)
 |    44 Mb/s,   61 us (2.13 ms)
 4 RTR-SUNET47.SLAC.Stanford.EDU (192.68.191.34)
                        -> 192.68.191.34 (1)           
 |   ?? b/s,   -19 us (2.08 ms)
 5?Core-gateway.Stanford.EDU (171.64.1.115)
 |    74 Mb/s,   -34 us (2.17 ms)
 6 i2-gateway.Stanford.EDU (171.64.1.209)
 |   412 Mb/s,   30 us (2.26 ms)
 7 STAN.POS.calren2.NET (171.64.1.213)
 |   180 Mb/s,   159 us (2.65 ms)
 8 SUNV--STAN.POS.calren2.net (198.32.249.73)
                        -> 198.32.249.73 (1)           
 |   230 Mb/s,   1.57 ms (5.84 ms)
 9?abilene--QSV.POS.calren2.net (198.32.249.62)
                        -> 198.32.249.62 (1)           
 |   406 Mb/s,   3.58 ms (13.0 ms)
10?losa-scrm.abilene.ucaid.edu (198.32.8.18)
 |   ?? b/s,   127 us (13.3 ms)
11 USC--abilene.ATM.calren2.net (198.32.248.85)
 |   324 Mb/s,   453 us (14.2 ms)
12 QAnh--USC.POS.calren2.net (198.32.248.17)
 |   187 Mb/s,   325 us (14.9 ms)
13 UCI--QAnh.POS.calren2.net (198.32.248.126)
                        -> 198.32.248.126 (1)           
 |   306 Mb/s,   621 us (16.2 ms)
14?UCR--UCI.POS.calren2.net (198.32.248.13)
 |    99 Mb/s,   779 us (17.9 ms)
15 CIT--UCR.POS.calren2.net (198.32.248.9)
 |   876 Mb/s,   -617 us (16.7 ms)
16 media-converter.caltech.edu (192.41.208.50)
                        -> 192.41.208.50 (1)           
 |    36 Mb/s,   353 us (17.7 ms)
17?SFL-border.dmz.caltech.edu (192.12.19.252)
 |    27 Mb/s,   89 us (18.3 ms)
18 Lauritsen-RSM.ilan.caltech.edu (131.215.254.112)

University of Colorado, Boulder Colorado, USA

For Colorado the bottlenecks appear to be between ESNET-A-GATEWAY and the ESnet router at LBL (28Mb/s) and at Colorado (20Mbps).

pathchar to tourmalet.Colorado.EDU (128.138.133.129)
 mtu limitted to 1500 bytes at FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 doing 32 probes at each of 64 to 1500 by 44
 0 host.SLAC.Stanford.EDU 
 |    25 Mb/s,   302 us (1.09 ms)
 1 router1.SLAC.Stanford.EDU 
 |    82 Mb/s,   169 us (1.57 ms)
 2 router2.SLAC.Stanford.EDU 
 |    98 Mb/s,   59 us (1.81 ms)
 3 router3.Stanford.EDU
 |   155 Mb/s,   -92 us (1.71 ms)
 4 ESNET-A-GATEWAY.SLAC.Stanford.EDU (192.68.191.18)
 |    28 Mb/s,   1.25 ms (4.64 ms)
 5 lbl2-atms.es.net (134.55.24.22)
 |   308 Mb/s,   34 us (4.75 ms)
 6 BERK--esnet.POS.calren2.net (198.32.249.57)
 |   279 Mb/s,   608 us (6.01 ms)
 7 SUNV--BERK.POS.calren2.net (198.32.249.14)
 |   507 Mb/s,   1.57 ms (9.18 ms)
 8 abilene--QSV.POS.calren2.net (198.32.249.62)
 |   127 Mb/s,   10.9 ms (31.0 ms)
 9 dnvr-scrm.abilene.ucaid.edu (198.32.8.2)
                        -> 198.32.8.2 (1)           
 |    69 Mb/s,   596 us (32.4 ms)
10?204.131.62.41 (204.131.62.41)
                        -> 204.131.62.41 (2)           
 |    85 Mb/s,   88 us (32.7 ms)
11?128.138.81.1 (128.138.81.1)
                        -> 128.138.81.1 (1)           
 |   1069 Mb/s,   25 us (32.7 ms)
12?128.138.81.194 (128.138.81.194)
 |   177 Mb/s,   -2 us (32.8 ms)
13 phys-engr-supa.Colorado.EDU (128.138.81.84)
 |    35 Mb/s,   176 us (33.5 ms)
14 johnson-133-gw.Colorado.EDU (128.138.145.133)
 |    20 Mb/s,   -13 us (34.1 ms)
15 tourmalet.Colorado.EDU (128.138.133.129)
15 hops, rtt 31.3 ms (34.1 ms), bottleneck  20 Mb/s, pipe 85711 bytes

The pathchar measured after the routing from SLAC to Colorado was changed to use Abilene/Internet 2 is shown below (measured on 9/20/00):

pathchar to denali.colorado.edu (128.138.133.18)
 mtu limitted to 1500 bytes at FLORA05.SLAC.Stanford.EDU (134.79.16.59)
 doing 32 probes at each of 64 to 1500 by 44
 0 FLORA05.SLAC.Stanford.EDU (134.79.16.59)
 |    24 Mb/s,   212 us (0.92 ms)
 1 RTR-CORE1.SLAC.Stanford.EDU (134.79.19.2)
 |    95 Mb/s,   178 us (1.40 ms)
 2 RTR-CGB6.SLAC.Stanford.EDU (134.79.135.6)
 |    85 Mb/s,   55 us (1.65 ms)
 3 RTR-DMZ.SLAC.Stanford.EDU (134.79.111.4)
                        -> 134.79.111.4 (1057)           
 |    48 Mb/s,   83 us (2.07 ms)
 4?RTR-SUNET47.SLAC.Stanford.EDU (192.68.191.34)
 |   2086 Mb/s,   -42 us (1.99 ms)
 5 Core-gateway.Stanford.EDU (171.64.1.115)
 |   119 Mb/s,   4 us (2.10 ms)
 6 i2-gateway.Stanford.EDU (171.64.1.209)
 |   145 Mb/s,   11 us (2.21 ms)
 7 STAN.POS.calren2.NET (171.64.1.213)
 |   340 Mb/s,   161 us (2.56 ms)
 8 SUNV--STAN.POS.calren2.net (198.32.249.73)
 |   195 Mb/s,   1.57 ms (5.77 ms)
 9 abilene--QSV.POS.calren2.net (198.32.249.62)
 |   ?? b/s,   10.9 ms (27.5 ms)
10 dnvr-scrm.abilene.ucaid.edu (198.32.8.2)
                        -> 198.32.8.2 (2)           
 |    43 Mb/s,   618 us (29.0 ms)
11?204.131.62.41 (204.131.62.41)
 |    98 Mb/s,   93 us (29.3 ms)
12 hut-cuatm.Colorado.EDU (128.138.81.1)
 |   ?? b/s,   12 us (29.3 ms)
13 engr-hut.Colorado.EDU (128.138.81.194)
 |   ?? b/s,   49 us (29.4 ms)
14 phys-engr-supb.Colorado.EDU (128.138.81.86)
 |    38 Mb/s,   163 us (30.0 ms),  +q 1.40 ms (6.60 KB) *5
15 johnson-133-gw.Colorado.EDU (128.138.145.133)
                        -> 128.138.145.133 (1)           
 |    17 Mb/s,   142 us (31.0 ms),  +q 1.33 ms (2.90 KB) *3,  14% dropped
16?denali.Colorado.EDU (128.138.133.18)
16 hops, rtt 28.3 ms (31.0 ms), bottleneck  17 Mb/s, pipe 67489 bytes

LLNL, Livermore California, USA

For LLNL the bottleneck is unclear. Intuitively one might expect it to be the SLAC ESnet ATM link since the other links are known to have greater capacity.

pathchar to south.llnl.gov (134.9.1.8)
 can't find path mtu.
 doing 32 probes at each of 64 to 4032 by 128
 0 host.SLAC.Stanford.EDU 
 |    38 Mb/s,   340 us (1.53 ms)
 1 router1.SLAC.Stanford.EDU
 |   ?? b/s,   215 us (1.95 ms)
 2 router2.SLAC.Stanford.EDU 
 |   269 Mb/s,   84 us (2.24 ms)
 3 router3.SLAC.Stanford.EDU
 |   ?? b/s,   -79 us (2.05 ms)
 4 ESNET-A-GATEWAY.SLAC.Stanford.EDU (192.68.191.18)
                        -> 192.68.191.18 (1)           
 |   164 Mb/s,   3.14 ms (8.53 ms)
 5?llnl3-atms.es.net (134.55.24.1)
                        -> 134.55.24.1 (1)           
 |   123 Mb/s,   125 us (9.05 ms)
 6?llnl-foutnet.llnl.gov (192.188.35.1)
 |    53 Mb/s,   308 us (10.3 ms)
 7 llnl-finternet.llnl.gov (128.115.248.2)
 |   119 Mb/s,   125 us (10.8 ms)
 8 olnfddi-ffastfddi.llnl.gov (128.115.237.2)
 9:  20  1088 392      11

Stanford University

The following is the pathchar from SLAC to Stanford University campus measured 9/20/00.

pathchar to elaine1.stanford.edu (171.64.15.66)
 mtu limitted to 1500 bytes at FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 doing 32 probes at each of 64 to 1500 by 44
 0 FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 |    26 Mb/s,   226 us (0.92 ms)
 1 RTR-CORE1.SLAC.Stanford.EDU (134.79.19.2)
 |    73 Mb/s,   169 us (1.42 ms)
 2 RTR-CGB6.SLAC.Stanford.EDU (134.79.135.6)
 |   106 Mb/s,   52 us (1.64 ms)
 3 RTR-DMZ.SLAC.Stanford.EDU (134.79.111.4)
                        -> 134.79.111.4 (2)           
 |    48 Mb/s,   87 us (2.06 ms)
 4?RTR-SUNET47.SLAC.Stanford.EDU (192.68.191.34)
 |   2398 Mb/s,   -35 us (2.00 ms)
 5 Core-gateway.Stanford.EDU (171.64.1.115)
 |    83 Mb/s,   -28 us (2.09 ms)
 6 i2-gateway.Stanford.EDU (171.64.1.209)
 |    49 Mb/s,   122 us (2.58 ms)
 7 Core4-gateway.Stanford.EDU (171.64.1.226)
 |    39 Mb/s,   95 us (3.07 ms)
 8 Core6-gateway.Stanford.EDU (171.64.3.81)
 |    47 Mb/s,   356 us (4.04 ms)
 9 171.64.1.230 (171.64.1.230)
10  * 1   504 533       4
11  * 1   372 531       45

Manchester University, Manchester, England

The pathchar from SLAC to Manchester Universisty in England indicates that there is a bottleneck of 7.5Mbps at Manchester. There also appears to be a 12 Mbps bottleneck between ESnet and DANTE at 60 Hudson Street in New York. There is also observed to be a lot of queueing after 60 Hudson Street.

pathchar to pc6.ph.man.ac.uk (130.88.20.21)
 mtu limitted to 1500 bytes at FLORA03.SLAC.Stanford.EDU (134.79.16.55)
 doing 32 probes at each of 64 to 1500 by 44
 0 host.SLAC.Stanford.EDU 
 |    25 Mb/s,   209 us (898 us)
 1 router1.SLAC.Stanford.EDU 
 |    75 Mb/s,   179 us (1.41 ms)
 2 router2.SLAC.Stanford.EDU 
                        ->  (3312)           
 |   121 Mb/s,   55 us (1.62 ms)
 3?router3.SLAC.Stanford.EDU (134.79.111.4)
 |   171 Mb/s,   -101 us (1.49 ms)
 4 ESNET-A-GATEWAY.SLAC.Stanford.EDU (192.68.191.18)
                        -> 192.68.191.18 (1)           
 |    26 Mb/s,   29.8 ms (61.5 ms)
 5?pppl4-atms.es.net (134.55.24.10)
 |    26 Mb/s,   1.09 ms (64.2 ms)
 6 60hudson-pppl.es.net (134.55.43.2)
                        -> 134.55.43.2 (3)           
 |    12 Mb/s,   262 us (65.7 ms),  +q 1.10 ms (1.59 KB) *5
 7?193.62.157.5 (193.62.157.5)
                        -> 193.62.157.5 (2)           
 |    25 Mb/s,   33.4 ms (133 ms),  +q 1.46 ms (4.61 KB) *4
 8?193.62.157.13 (193.62.157.13)
                        -> 193.62.157.13 (1)           
 |    36 Mb/s,   -13 us (133 ms),  +q 1.41 ms (6.33 KB) *4
 9?193.63.94.246 (193.63.94.246)
                        -> 193.63.94.246 (1)           
 |   ?? b/s,   4.27 ms (142 ms)
10?193.62.157.82 (193.62.157.82)
 |    31 Mb/s,   3.23 ms (149 ms),  +q 1.84 ms (7.04 KB) *4
11 146.97.255.178 (146.97.255.178)
                        -> 146.97.255.178 (2)           
 |    68 Mb/s,   -2194 us (144 ms),  +q 3.01 ms (25.4 KB) *2
12?gw-owens.g-ming.net.uk (194.66.23.2)
 |    42 Mb/s,   -213 us (144 ms),  +q 4.51 ms (23.5 KB) *2
13 130.88.253.1 (130.88.253.1)
 |   7.5 Mb/s,   9.70 ms (165 ms),  +q 3.38 ms (3.18 KB) *2
14 pc6.ph.man.ac.uk (130.88.20.21)
14 hops, rtt 159 ms (165 ms), bottleneck 7.5 Mb/s, pipe 155000 bytes

INFN, Rome Italy

The pathchar to INFN/Roma shows a 7Mb/s bottlneck between GARR and Roma.

pathchar to bbfarm01.roma1.infn.it (141.108.23.11)
 mtu limitted to 1500 bytes at FLORA02.SLAC.Stanford.EDU (134.79.16.57)
 doing 32 probes at each of 64 to 1500 by 44
 0 host.SLAC.Stanford.Edu
 |    25 Mb/s,   214 us (0.90 ms)
 1 router1.SLAC.Stanford.Edu
 |    73 Mb/s,   175 us (1.41 ms)
 2 router2.SLAC.stanford.Edu
 |   106 Mb/s,   63 us (1.66 ms)
 3 router3.SLAC.Stanford.Edu
 |   173 Mb/s,   -99 us (1.53 ms)
 4 ESNET-A-GATEWAY.SLAC.Stanford.EDU (192.68.191.18)
                        -> 192.68.191.18 (2)           
 |    26 Mb/s,   29.8 ms (61.6 ms)
 5?pppl4-atms.es.net (134.55.24.10)
                        -> 134.55.24.10 (3)           
 |    26 Mb/s,   1.10 ms (64.3 ms)
 6?60hudson-pppl.es.net (134.55.43.2)
                        -> 134.55.43.2 (2)           
 |    24 Mb/s,   398 us (65.6 ms)
 7?esnet.ny.dante.net (212.1.200.217)
 |    43 Mb/s,   51.0 ms (168 ms)
 8 212.1.200.18 (212.1.200.18)
 |    28 Mb/s,   4.18 ms (177 ms)
 9 rm-mi-2.garr.net (193.206.134.22)
                        -> 193.206.134.162 (14016)           
                        -> 193.206.134.22 (4)       
 |    73 Mb/s,   312 us (177 ms)
10?rc-rt-2.rm.garr.net (193.206.134.166)
 |   7.2 Mb/s,   662 us (180 ms)
11 infnrmI-rc.rm.garr.net (193.206.131.14)
                        -> 193.206.131.14 (3)           
 |   172 Mb/s,   -148 us (180 ms),  +q 1.06 ms (22.7 KB) *2
12?bbfarm01.roma1.infn.it (141.108.23.11)
12 hops, rtt 175 ms (180 ms), bottleneck 7.2 Mb/s, pipe 162084 bytes

Daresbury Lab, nr. Liverpool England

The pathchar from SLAC to the Daresbury Lab near Liverpool, England shows a bottleneck of 10Mbits/s, a pipe of 190kBytes, and queuing between hops 7 and 8, 9 and 10 (the traceroute identifes that these are in AS 786, hops 7 and 8 are identified as JANET, hop 9 is the University of London Computer Center, hops 10 and 11 are the UK Academic Joint Network Team).

pathchar to srgen1.dl.ac.uk (148.79.240.5)
 mtu limitted to 1500 bytes at FLORA01.SLAC.Stanford.EDU (134.79.16.29)
 doing 32 probes at each of 64 to 1500 by 44
 0 host.SLAC.Stanford.EDU
 |    25 Mb/s,   219 us (0.93 ms)
 1 router1.SLAC.Stanford.EDU
 |    80 Mb/s,   170 us (1.42 ms)
 2 router2.SLAC.Stanford.EDU (134.79.x.x)
                        -> 134.79.x.x (275213)           
 |   107 Mb/s,   55 us (1.64 ms)
 3?router3.SLAC.Stanford.EDU
 |   116 Mb/s,   -100 us (1.54 ms)
 4 ESNET-A-GATEWAY.SLAC.Stanford.EDU (192.68.191.18)
 |    27 Mb/s,   29.8 ms (61.6 ms)
 5 pppl4-atms.es.net (134.55.24.10)
                        -> 134.55.24.10 (2)           
 |    26 Mb/s,   1.07 ms (64.2 ms)
 6?60hudson-pppl.es.net (134.55.43.2)
 |    10 Mb/s,   184 us (65.7 ms)
 7 193.62.157.5 (193.62.157.5)
                        -> 193.62.157.5 (2)           
 |    31 Mb/s,   33.2 ms (133 ms),  +q 1.36 ms (5.20 KB) *2
 8?193.62.157.17 (193.62.157.17)
                        -> 193.62.157.17 (1)           
 |   ?? b/s,   763 us (134 ms)
 9?128.86.1.241 (128.86.1.241)
                        -> 128.86.1.241 (3)           
 |    14 Mb/s,   7.31 ms (149 ms),  +q 5.66 ms (9.90 KB)
10?146.97.253.173 (146.97.253.173)
                        -> 146.97.253.173 (1)           
 |    44 Mb/s,   -2978 us (143 ms),  +q 6.22 ms (34.5 KB),  1% dropped
11?146.97.253.178 (146.97.253.178)
 |   ?? b/s,   471 us (144 ms),  1% dropped
12 gw-dl.netnw.net.uk (194.66.24.2)
                        -> 194.66.24.2 (1)           
 |    11 Mb/s,   3.87 ms (153 ms),  +q 6.02 ms (8.49 KB),  1% dropped
13?gw-fw.dl.ac.uk (193.63.74.233)
                        -> 193.63.74.233 (2)           
 |   ?? b/s,   -3865 us (145 ms),  1% dropped
14?alan3.dl.ac.uk (193.63.74.129)
                        -> 193.63.74.129 (2)           
 |    46 Mb/s,   399 us (146 ms),  +q 4.80 ms (27.5 KB),  8% dropped
15?srgen1.dl.ac.uk (148.79.240.5)
15 hops, rtt 141 ms (146 ms), bottleneck  10 Mb/s, pipe 190175 bytes

Created August25, 2000
Comments to iepm-l@slac.stanford.edu