The PingER MA sends up to 30 x 100Byte pings at 1 second intervals (or 800bits/second) and up to 30 1000Byte pings at 1 second intervals (8000/second) to each target until it receives up to 10 responses from the target or times out in 30 seconds. In the case of the main MA at SLAC that monitors about 800 nodes, each 30 minutes this corresponds to an aggregate data rate of ~7Mbit/sec.
Estimating the throughput using the ping loss and Round Trip Time (RTT) and the Mathis formula (see https://www.slac.stanford.edu/comp/net/wan-mon/thru-vs-loss.html) and comparing with the TCP throughput using ttcp (see https://en.wikipedia.org/wiki/Ttcp) gave good agreement.
This uses The macroscopic behavior of the TCP congestion avoidance algorithm by Mathis, Semke, Mahdavi & Ott in Computer Communication Review, 27(3), July 1997, that provides a short and useful formula for the upper bound on the transfer rate: Rate <= (MSS/RTT)*(1 / sqrt{p})
where:
Rate: is the TCP transfer rate or throughput
MSS: is the maximum segment size (fixed for each Internet path, typically 1460 bytes)
RTT: is the round trip time (as measured by TCP)
p: is the packet loss rate.
This is similar to monitor-io (see https://www.monitor-io.com/monitor-io-vs-speed-tests.html), where it says:
The following are some providers of measurements. They use different techniques, are made at different times so comparisons are questionable. If for example, we choose to compare the results for S. E. Asian countries we get
Nation | Wikipedia (2017/Q1) Mbps | Atlas and Boots (2018) Mbps | Speedtest (7/2019) Mbps | Global Economy (2016) Mbps | Testmy.net upload (2019) Mbps | PingER (2019 Jan-Sep) Mbps |
---|---|---|---|---|---|---|
Brunei | 4.78 (8) | 20.97 (6) | 3.7 (9) | 0.681(8) | ||
Cambodia | 4.03 (9) | 18.51 (9) | 0.0061 (6) | 6.7 (8) | 0.918 (6) | |
Indonesia | 6.65 (5) | 19.09 (8) | 0.0062 (5) | 7.4 (6) | 0.910 (7) | |
Laos | 2.92 (10) | 19.27 (7) | 0.0028 (7) | 0.277(10) | ||
Malaysia | 8.9 (4) | 23.86 (2) | 71.52 (3) | 0.0271 (4) | 13.5 (4) | 1.429 (2) |
Myanmar | 4.87 (7) | 16.84 (10) | 0.0028 (7) | 6.9 (7) | 0.981 (5) | |
Philippines | 5.5 (5) | 6.05 (6) | 23.6 (5) | 0.0277 (3) | 8.1 (5) | 1.131 (4) |
Singapore | 20.3 (1) | 70.36 (1) | 197 (1) | 0.617 (1) | 17.2 (3) | 1.793 (1) |
Thailand | 16 (2) | 18.21 (3) | 78.47 (2) | 0.0548 (2) | 19.9 (1) | 1.175 (3) |
Vietnam | 9.5 (3) | 7.02 (4) | 38.63 (4) | 0.0208 (9) | 17.5 (2) | 0.681 (8) |
While the absolute values differ by orders of magnitude, the relative positions (in parentheses) are more stable. We can get a quantitative measure of the correlation between the various measurements and PingER by plotting each value against the PIngER value and calculating the R2 which is a measure of how much of the variance in y is explained by the model x (or in our case PingER). We do this for the measurements that have >=6 data points (ASEAN nations) to compare.
Measurement by | R2 |
---|---|
Atlas and Boots | 0.61 |
Speedtest | 0.47 |
Global Economy | 0.77 |
The yearly time series of the PingER measured derived throughputs is shown below:
https://www.speedtest.net/global-index. See above for the R2 comparisons.
July 2019
Provides both mobile and fixed line measurements. Based on measurements made to subscribers.
Test your Internet connection bandwidth to locations around the world with this interactive broadband speed test from Ookla.
https://www.fastmetrics.com/internet-connection-speed-by-country.php#top-10-comparison-2018
Top 50 average download speeds June 2017 - May 2018. There are only 3 ASEAN nations in FastMetrics so it is not used in the R2 comparisons.
Latest data 2017 Q1.
https://en.wikipedia.org/wiki/List_of_countries_by_Internet_connection_speeds. There is insufficient data (number 5 is < cutoff 6) of ASEAN countries) in Wikipedia so it is not used in the R2 comparison.
https://www.atlasandboots.com/remote-jobs/countries-with-the-fastest-internet-in-the-world/. See above for the R2 comparisons.
Countries with fastest Internet in 2018
See https://www.monitor-io.com/monitor-io-vs-speed-tests.html. Costs money. Unable to find any results ordered by country. Not reported above.
https://www.checkmyspeednow.com/ makes measurements to subscribers. Unable to find any results reported by country. Not reported above.
https://www.thinkbroadband.com/broadband/monitoring/quality. Can't find data of throughput by country. Not reported above.
The line monitoring system allows you to track the performance of your broadband connection in terms of latency and packet loss. Latency is the time it takes for a piece of information (a 'packet') to get from our servers to your broadband connection, and back. Packet loss is where some bits of information are lost whilst in transit, so it has to be re-sent. These can cause problems with various 'realtime' applications such as online gaming, video conferencing, internet telephony, etc.
https://www.theglobaleconomy/com/rankings/Internet_bandwidth/. See above for the R2 comparisons.
Based on measurements made to subscribers. Provides upload and download. We use the upload. The results are strange in that all the other measurements, Singapore is ranked behind Thailand and Vietnam. Thus we not use this data in the comparisons.
Their most recent data is for 2005 for which PingER has data for only 4 ASEAN nations (Brunei, Indonesia, Malaysia and Singapore). Thus we do not use this data for comparisons.
See PingER and perfSONAR comparison.