Estimating speedtest server accuracy
Edge clusters execute in a plurality of regional clouds of a cloud computing platforms, which may include cloud POPs. Edge clusters may be programmed to control access to applications executing in the cloud computing platform. Edge clusters and an intelligent routing module route traffic to applications executing in the cloud computing platform. Cost and latency may be managed by the intelligent routing module by routing requests over the Internet or a cloud backbone network and using or bypassing cloud POPs. The placement of edge clusters may be selected according to measured or estimated latency. Latency may be estimated using speed test servers and the locations of speed test servers may be verified.
1. A method comprising:
obtaining, by a computing module, for each speed test server of a plurality of speed test servers, measurements of latency between each speed test server and a plurality of geographically distributed locations, each speed test server having a location associated therewith;
for each speed test server, evaluating, by the computing module, the measurements of latency between each speed test server and the plurality of geographically distributed locations with respect to the location associated with each speed test server; and
determining, by the computing module, that a first speed test server of the plurality of speed test servers is suspect according to the evaluating of the measurements of latency between the first speed test server and the plurality of geographically distributed locations;
comparing first measurements of latency between the first speed test server and the plurality of geographically distributed locations to one or more second measurements of latency between a second speed test server and the plurality of geographic locations, a first location being associated with the first speed test server and a second location being associated with the second speed test server;
determining that a difference between the first measurements and the second measurements does not correspond to a distance between the first location and the second location; and
in response to determining that the difference between the first measurements and the second measurements does not correspond to the distance between the first location and the second location, determining that the first speed test server is suspect and ignoring measurements of latency from the first speed test server.
2. The method of claim 1 , further comprising:
determining that a first measurement of latency between the first speed test server and a first location of the plurality of geographically distributed locations is below a baseline latency; and
in response to determining that the first measurement of latency is below the baseline latency, determining that the first speed test server is suspect.
3. The method of claim 2 , wherein the baseline latency is a traversal time between the first location and the first speed test server at the speed of light.
4. The method of claim 2 , wherein the baseline latency is a traversal time over a segment of network infrastructure between the first speed test server and the first location.
5. The method of claim 4 , wherein the segment of network architecture is a cloud backbone network connecting a first regional cloud and a second regional cloud of a cloud computing platform, the first speed test server being in a first geographic region associated with the first regional cloud and the first location being a computing module executing in the second regional cloud.
6. The method of claim 1 , further comprising:
grouping the plurality of speed test servers into a plurality of clusters according to similarity of the measurements of latency between each speed test server and the plurality of geographically distributed locations.
7. The method of claim 6 , further comprising:
determining (a) that a second location associated with a second speed test server of the plurality of speed test servers does not correspond to third locations associated with other speed test servers of the plurality of speed test servers in a cluster of the plurality of clusters including the second speed test server; and
in response to (a), associating a corrected location with the second speed test server, the corrected location corresponding to the third locations.
8. The method of claim 1 , wherein measurements of latency between each speed test server and a plurality of geographically distributed locations comprise measurements between each speed test server and a plurality of edge clusters executing within a plurality of regional clouds of a cloud computing platform.
9. The method of claim 8 , further comprising:
selecting, by the computing module, regional clouds of the plurality of regional clouds for instantiating each edge cluster of the plurality of edge clusters according to the measurements of latency for the plurality of speed test servers excluding the first speed test server.