...
In this report the jitter is represented by the Inter Packet Delay (IPD), the absolute values of the IPF (Abs(IPD)) and the Stdev(IPD).
Measurements
Methodology
Since we believed, both a priori and from observations, that the major impact on the measurements was the network and not the servers' hardware or OS, we We chose to make detailed measurements to and from two analysis of PingER measurements between two representative MAs at SLAC and two sites well separated from SLAC and hence with very different RTTs. The MAs at SLAC .were:
The Dell Poweredge 2650 bare metal pinger.slac.stanford.edu server running Red Hat Linux 2.6.32-504.8.1.el6.i686 with specs.
Code Block 103cottrell@pinger:~$uname -a Linux pinger 2.6.32-504.8.1.el6.i686 #1 SMP Fri Dec 19 12:14:17 EST 2014 i686 i686 i386 GNU/Linux
The Raspberry Pi pinger-raspberry.slac.stanford.edu an armv61 running Gnu Linux (see above) with specs.
Code Block pi@pinger-raspberry ~ $ uname -a Linux pinger-raspberry.slac.stanford.edu 3.18.11+ #781 PREEMPT Tue Apr 21 18:02:18 BST 2015 armv6l GNU/Linux
Both were in the same building at SLAC, i.e. roughly at latitude 37.4190 N, longitude 122.2085 W, but on different floors. The machines are about 30 metres apart or about 0.0003 msec based on the speed of light in a direct fibre.
The measurements used were made:
- between pinger.slac.stanford.edu and pinger-raspberry.slac.stanford.edu, and
- from both pinger.slac.stanford.edu and pinger-raspberry.slac.stanford.edu to targets at varying distances and hence varying minimum RTTs from SLAC.
- from two representative MAs at sitka.triumf.ca and ping.cern.ch to pinger.slac.stanford.edu and pinger-raspberry.edu
...
- The Min RTT error is the IQR of the minimum RTTs from each set of pings for the time period.
- The Avg RTT error is the standard deviation of all the average RTTs from each set of pings for the time period
- The Median RTT error is the IQR of all the individual RTTs from all sets of measurements for the time period
- The IQR IPD error is the IQR of the hourly IQRs for the time period (see Spreadsheet)
Metric \ Monitor Target | pinger to pinger-raspberry | pinger-raspberry to pinger | pinger to sitka | pinger-raspberry to sitka | sitka to pinger | sitka to pinger-raspberry | pinger to CERN | pinger-raspberry to CERN |
---|---|---|---|---|---|---|---|---|
Time period | June 17 - July 13, 2015 | June 17 - July 13, 2015 | June 17 - July 14, 2015 | June 17 - July 14 | July 15 - July 16 2015 | July 15 - July 16 2015 | June 17 - July 12, 2015 | June 17 - July 12, 2015 |
Samples | 10890 | 10850 | 10900 | 10850 | 820 | 820 | 10900 | 10900 |
Min RTT | 0.43 ms | 0.41 ms | 22 +- 1 ms | 22.3 +- 1 ms | 22 ms | 22.4 ms | 150 +- 1 ms | 151 ms |
Avg RTT | 0.542 +- 0.06 ms | 0.529 +- 0.5ms | 23.9 +- 1ms | 23.827 +- 1 ms | 22.30 ms | 22.709 ms | 150.307 +- 2.4 ms | 151.024 +- 1.3 ms |
Max RTT | 1.15 ms | 20.8 ms | 761 ms | 334 ms | 26.3 ms | 25.9 ms | 288 ms | 183 ms |
Stdev | 0.055 ms | 0.540 ms | 9.58 ms | 9.60 ms | 0.219 ms | 0.222 ms | 2.37 ms | 1.3 ms |
Median RTT | 0.542 +- 0.05 ms | 0.51 ms +- 0.052 ms | 22.3 +- 0.2 ms | 22.7 +- 0.11 ms | 22.3 +- 0.11 ms | 22.7 +- 0.21 ms | 150 +- 1 ms | 151 +- 0.01 ms |
25% | 0.514 ms | 0.48 ms | 22.2 ms | 22.69 ms | 22.3 ms | 22.8 ms | 149.99 ms | 150.99 ms |
75% | 0.564 ms | 0.532 ms | 22.4 ms | 22.8 ms | 22.19 ms | 22.59 ms | 151 ms | 151 ms |
IQR | 0.05 ms | 0.052 ms | 0.2 ms | 0.11 ms | 0.113 ms | 0.210 ms | 1.01 ms | 0.01 ms |
Min IPD | -0.59 ms | -20.29 ms | -244 ms | -235.1 ms | -0.4 ms | -0.39 ms | -137 ms | -32 ms |
Avg IPD | 0 ms | 0 ms | 0 ms | 0 ms | 0 ms | 0 ms | 0 ms | 0 |
Max IPD | 0.62 ms | 20.26 ms | 268 ms | 146 ms | 3.6 ms | 3.29 ms | 138 ms | 32 ms |
Median IPD | 0 +- 0.07 ms | 0 +- 0.05 ms | 0 +- 0.2 ms | 0 +- 0.2ms | 0 ms +- 0.2 ms | 0 +- 0.2 ms | 0 +- 0.01 ms | 0 +- 0.01 ms |
Stdev | 0.248 ms | 0.296 ms | ||||||
25% IPD | -0.04 ms | -0.03 ms | -0.1 ms | -0.11 ms | -0.1 ms | -0.1 ms | -0.01 ms | -0.01 ms |
75% IPD | 0.03 ms | 0.02 ms | 0.09 ms | 0.1 ms | 0.01 ms | 0.09 ms | 0 ms | 0 ms |
IQR IPD | 0.07 ms | 0.05 ms | 0.190 +- 0.5 ms | 0.210 +- 0.4 ms | 0.2 ms | 0.190ms | 0.01 ms | 0.01 ms |
Min(abs(IPD)) | 0 ms | 0 ms | 0 ms | 0 ms | 0 ms | 0 ms | 0 ms | 0 ms |
Avg(abs(IPD)) | 0.041 ms | 0.066 ms | 0.424 ms | 0.406 ms | 0.120 ms | 0.139 ms | 0.386 ms | 0.046 ms |
Max(abs(IPD)) | 0.0628ms | 20.294 ms | 268 ms | 235.1 ms | 4 ms | 3.29 ms | 138 ms | 32 ms |
Stdev | 0.248 ms | 0.262 ms | ||||||
Median(abs(IPD)) | 0.03ms | 0.024 ms | 0.09 ms | 0.1 ms | 0.1 ms | 0.09 ms | 0 ms | 0 ms |
25%(abs(IPD)) | 0.01ms | 0.008 ms | 0.01 ms | 0.01 ms | -0.01 ms | -0.01 ms | 0.01 ms | 0.01 ms |
75%(abs(IPD)) | 0.058 ms | 0.05 ms | 0.1 ms | 0.19 ms | 0.1 ms | 0.19 ms | 1 ms | 0 ms |
IQR(abs(IPD) | 0.048 ms | 0.042 ms | 0.09 ms | 0.11 ms | 0.11 ms | 0.2 ms | 0.009 ms | 0.01 ms |
Loss | 0% | 0% | 0% | 0.008% | 0% | 0% | 0% | 0% |
To find the probability of the distributions overlapping use a nomogram of mean differences versus error ratios given in Overlapping Normal Distributions. John M. Linacre for normal distributions.
We also use the Z Tests
Spreadsheet
Kolmogorov-Smirnov Test
The Kolmogorov-Smirnov test (KS-test) tries to determine if two datasets differ significantly. The KS-test has the advantage of making no assumption about the distribution of data. In other words it is non-parametric and distribution free. The method is explained here and makes use of an Excel tool called "Real Statiscs". The tests were made using the raw data and distributions, both methods had similar results except for the 100Bytes Packet that had a great difference in the results. The results using raw data says both samples does not come from the same distribution with a significant difference, however if we use distributions the result says that only the 1000Bytes packet does not come from the same distribution. Bellow you will find the graphs for the distributions that were created and the cumulative frequency in both cases plotted one above other (in order to see the difference between the distributions).
...