Network optimization system using server latency measurements

1. A method, comprising:

receiving, at a first server, a first request from a client device;

determining that a latency measurement instruction has been issued;

based on determining that the latency measurement instruction has been issued and in response to receiving the first request from the client device, providing a first redirect response to the client device, the first redirect response causing the client device to submit a second request to a second server;

receiving, at the first server, a final request, the final request being provided to the first server from the client device based on the client device receiving a second redirect response from the second server;

determining a latency measurement between the client device and the second server based, at least in part, on a time between when the first redirect response is provided to the client device and when the final request is received by the first server;

determining whether the latency measurement is above a latency threshold; and

in response to determining that the latency measurement is above the latency threshold, causing the client device to access the second server instead of the first server.

2. The method of claim 1 , further comprising providing the latency measurement to a coordination system.

3. The method of claim 1 , wherein the second server provides the second redirect response to the client device in response to the second server receiving the second request.

4. The method of claim 1 , wherein the first request is provided to the first server based, at least in part, on a property associated with the client device.

5. The method of claim 4 , wherein the property indicates an amount of time the client device has been sending requests to the first server.

6. The method of claim 1 , wherein the first server stores information associated with the second server.

7. The method of claim 6 , wherein the information associated with the second server comprises a hop count associated with the second server.

8. A system, comprising:

a processor; and

a memory coupled to the processor and storing instructions that, when executed by the processor, perform operations, comprising:

receiving a first request from a client device;

receiving a latency measurement instruction;

based on receiving the latency measurement instruction and in response to receiving the first request from the client device, providing a first redirect response to the client device, the first redirect response causing the client device to submit a second request to a server;

receiving a final request, the final request being provided from the client device based on the client device receiving a second redirect response from the server;

determining a latency measurement between the client device and the server based, at least in part, on a time between when the first redirect response is provided to the client device and when the final request is received;

determining whether the latency measurement is above a latency threshold; and

in response to determining that the latency measurement is above the latency threshold, causing the client device to access the server.

9. The system of claim 8 , further comprising instructions for providing the latency measurement to a coordination system.

10. The system of claim 8 , wherein the server is not visible to the client device.

11. The system of claim 8 , wherein the first request is received based, at least in part, on a property associated with the client device.

12. The system of claim 11 , wherein the property indicates an amount of time the client device has been sending requests.

13. The system of claim 8 , further comprising instructions for storing information associated with the server.

14. The system of claim 13 , wherein the information associated with the server comprises a hop count associated with the server.

15. A method, comprising:

providing a latency measurement instruction to a first server, the latency measurement instruction causing:

the first server to receive a first request from a client device;

the first server to provide a first redirect response to the client device, the first redirect response causing the client device to submit a second request to a second server;

the first server to receive a final request, the final request being provided from the client device based on the client device receiving a second redirect response from the second server;

the first server determining a latency measurement between the client device and the first server based, at least in part, on a time associated with the first redirect response and the final request; and

the first server determining whether the latency measurement is above a latency threshold; and

receiving an instruction from the first server causing the client device to access the second server instead of the first server in response to the first server determining that the latency measurement is above the latency threshold.