Estimation of latency across regional clouds
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:
estimating, by a computing module, L1 latency between a user endpoint and a first regional cloud R1 of a first cloud computing platform, the first regional cloud R1 corresponding to a first geographic region and the user endpoint being located in a second geographic region different from the first geographic region, the second geographic region associated with a second regional cloud R2 of the first cloud computing platform, the first regional cloud R1 and the second regional cloud R2 being connected to one another by a cloud backbone network the user endpoint being connected to the second regional cloud R2 by a wide area network WAN that is external to the first cloud computing platform and does not include the cloud backbone network, the estimating L1 latency comprising:
identifying a secondary regional cloud S2 of a second cloud computing platform different from the first cloud computing platform, the secondary regional cloud S2 corresponding to the second geographic region;
measuring an L2 latency between the first regional cloud R1 and a secondary regional cloud S2 of a second cloud computing platform different from the first cloud computing platform;
obtaining a local latency estimate for the user endpoint, the local latency estimate corresponding to latency of the WAN within the second geographic region;
generating the L1 latency as a combination of the L2 latency and the local latency estimate;
providing an application instance executing in the first regional cloud R1;
providing an edge cluster executing in the first regional cloud R1, the edge cluster programmed to control access to the application instance;
selecting, by the computing module, a path to the application instance according to the L1 latency;
programming, by the computing module, the first cloud computing platform to route traffic from the user endpoint to the application instance according to the path; and
routing the traffic from the user endpoint to the application instance according to the path through one or both of the WAN and the cloud backbone network.
2. The method of claim 1 , wherein measuring the L2 latency comprises measuring the L2 latency using cloud census agents executing in the first regional cloud R1 and the secondary regional cloud S2.
3. The method of claim 1 , wherein the first cloud computing platform and the second cloud computing platform are different entities selected from the group consisting of AMAZON WEB SERVICES (AWS), MICROSOFT AZURE, and GOOGLE CLOUD PLATFORM (GCP).
4. The method of claim 1 , wherein the WAN is the Internet.
5. The method of claim 1 , wherein obtaining the local latency estimate comprises estimating a network traversal time between the user endpoint and the secondary regional cloud S2 based on measurements of network speed within the WAN in the second geographic region.
6. The method of claim 1 , further comprising:
measuring actual latency between the user endpoint and the first regional cloud R1; and
updating the L1 latency according to the actual latency.
7. The method of claim 1 , wherein selecting the path comprises, by the computing module:
selecting between the WAN and the cloud backbone network to convey traffic from the second geographic region to the first regional cloud R1.
8. A system comprising one or more processing devices and one or more memory devices operably coupled to the one or more processing devices, the one or more memory devices storing executable code that, when executed by the one or more processing devices, causes the one or more processing devices to:
estimate L1 latency between a user endpoint and a first regional cloud R1 of a first cloud computing platform, the first regional cloud R1 corresponding to a first geographic region and the user endpoint being located in a second geographic region different from the first geographic region, the second geographic region associated with a second regional cloud R2 of the first cloud computing platform, the first regional cloud R1 and the second regional cloud R2 being connected to one another by a cloud backbone network, the user endpoint and the second regional cloud R2 being connected to one another by a wide area network WAN that is external to the first cloud computing platform and does not include the cloud backbone network, by:
identifying a secondary regional cloud S2 of a second cloud computing platform different from the first cloud computing platform, the secondary regional cloud S2 corresponding to the second geographic region;
measuring an L2 latency between the first regional cloud R1 and a secondary regional cloud S2 of a second cloud computing platform different from the first cloud computing platform;
obtaining a local latency estimate for the user endpoint, the local latency estimate corresponding to latency of the WAN within the second geographic region; and
generating the L1 latency as a combination of the L2 latency and the local latency estimate;
wherein the executable code, when executed by the one or more processing devices, further causes the one or more processing devices to:
instantiate an application instance executing in the first regional cloud R1; and
instantiate an edge cluster executing in the first regional cloud R1, the edge cluster programmed to control access to the application instance; and
wherein the executable code, when executed by the one or more processing devices, further causes the one or more processing devices to:
select a path to the application instance according to the L1 latency;
program a domain name service (DNS) of the first cloud computing platform to route traffic from the user endpoint to the application instance according to the path; and
route the traffic from the user endpoint to the application instance according to the path through one or both of the WAN and the cloud backbone network.
9. The system of claim 8 , wherein measuring the L2 latency comprises measuring the L2 latency using cloud census agents executing in the first regional cloud R1 and the secondary regional cloud S2.
10. The system of claim 8 , wherein the first cloud computing platform and the second cloud computing platform are different entities selected from the group consisting of AMAZON WEB SERVICES (AWS), MICROSOFT AZURE, and GOOGLE CLOUD PLATFORM (GCP) and the WAN is the Internet.
11. The system of claim 8 , wherein obtaining the local latency estimate comprises estimating a network traversal time between the user endpoint and the secondary regional cloud S2 based on measurements of network speed within the WAN in the second geographic region.
12. The system of claim 8 , wherein the executable code, when executed by the one or more processing devices, further causes the one or more processing devices to:
measure actual latency between the user endpoint and the first regional cloud R1; and
update the L1 latency according to the actual latency.
13. The system of claim 8 , wherein the executable code, when executed by the one or more processing devices, further causes the one or more processing devices to select the path by selecting between the WAN and the cloud backbone network to convey traffic from the second geographic region to the first regional cloud R1.