Granted Patent B2

US 12,244,682 · App. 18/463,639 · Granted Mar 4, 2025

Inventor: Ian Wilkes (San Francisco, CA)

Assignee: Hound Technology, Inc.

H04L67/60H04L43/0823H04L43/16

Abstract

Claims (50)

1. A computer-implemented method comprising:

determining, by a stateful server, a first set of requests in association with a client operation;

issuing, by the stateful server, the first set of requests to a stateless computing platform;

determining, by the stateful server, whether a cumulative percentage of requests serviced with a response from the stateless computing platform in the first set of requests issued to the stateless computing platform meets a threshold;

responsive to determining that the cumulative percentage of requests serviced with the response from the stateless computing platform in the first set of requests issued to the stateless computing platform meets the threshold, reissuing, by the stateful server, a remainder of the requests from the first set of requests as a second set of requests to the stateless computing platform, the remainder of the requests pending a response from the stateless computing platform prior to the reissuing; and

monitoring, by the stateful server, the first set and the second set of requests in parallel until the remainder of the requests in either the first set or the second set of requests is serviced with a response from the stateless computing platform.

2. The computer-implemented method of claim 1 , wherein reissuing the remainder of the requests from the first set of requests as the second set of requests to the stateless computing platform further comprises:

identifying the remainder of the requests in the first set of requests as having a potential issue;

copying the remainder of the requests from the first set of requests to the second set of requests; and

issuing the second set of requests to the stateless computing platform.

3. The computer-implemented method of claim 1 , wherein reissuing the remainder of the requests from the first set of requests as the second set of requests to the stateless computing platform further comprises retaining the remainder of the requests in the first set of requests and waiting for the remainder of the requests in the first set of requests to be serviced by the stateless computing platform.

4. The computer-implemented method of claim 1 , wherein monitoring the first set and the second set of requests in parallel until the remainder of the requests in either the first set or the second set of requests is serviced with a response from the stateless computing platform further comprises:

identifying whether the remainder of the requests in one of the first set and the second set of requests is serviced before other one of the first set and the second set of requests by the stateless computing platform; and

responsive to identifying that the remainder of the requests in one of the first set and the second set of requests is serviced before other one of the first set and the second set of requests, canceling the remainder of the requests in the other one of the first set and the second set of requests issued to the stateless computing platform.

5. The computer-implemented method of claim 1 , further comprising:

determining whether servicing of the cumulative percentage of requests in the first set of requests by the stateless computing platform to meet the threshold is possible; and

responsive to determining that the servicing of the cumulative percentage of requests in the first set of requests by the stateless computing platform to meet the threshold is impossible, reissuing at least a portion of the first set of requests to the stateless computing platform based on a tradeoff between a cost of reissue and servicing of the first set of requests.

6. The computer-implemented method of claim 1 , further comprising:

determining a failure rate of the stateless computing platform; and

determining the threshold based on the failure rate of the stateless computing platform.

7. The computer-implemented method of claim 6 , wherein the threshold is a difference between 100 and a multiple of the failure rate.

8. The computer-implemented method of claim 6 , wherein a number of times that the second set of requests is reissued to the stateless computing platform is based on the failure rate.

9. The computer-implemented method of claim 6 , wherein the failure rate is one of a statistical failure rate and a dynamically measured failure rate.

10. The computer-implemented method of claim 1 , wherein the threshold is set in a range from 85 percent to 99 percent.

11. A system comprising:

a stateless computing platform configured to receive and service a first set of requests and a second set of requests from a stateful server;

the stateful server configured to:

determine the first set of requests in association with a client operation;

issue the first set of requests to the stateless computing platform;

determine whether a cumulative percentage of requests serviced with a response from the stateless computing platform in the first set of requests issued to the stateless computing platform meets a threshold;

responsive to determining that the cumulative percentage of requests serviced with the response from the stateless computing platform in the first set of requests issued to the stateless computing platform meets the threshold, reissue a remainder of the requests from the first set of requests as the second set of requests to the stateless computing platform, the remainder of the requests pending a response from the stateless computing platform prior to the reissue; and

monitor, in parallel, the first set and the second set of requests until the remainder of the requests in either the first set or the second set of requests is serviced with a response from the stateless computing platform.

12. The system of claim 11 , wherein to reissue the remainder of the requests from the first set of requests as the second set of requests to the stateless computing platform, the stateful server is further configured to:

identify the remainder of the requests in the first set of requests as having a potential issue;

copy the remainder of the requests from the first set of requests to the second set of requests; and

issue the second set of requests to the stateless computing platform.

13. The system of claim 11 , wherein to reissue the remainder of the requests from the first set of requests as the second set of requests to the stateless computing platform, the stateful server is further configured to retain the remainder of the requests in the first set of requests and wait for the remainder of the requests in the first set of requests to be serviced by the stateless computing platform.

14. The system of claim 11 , wherein to monitor the first set and the second set of requests in parallel until the remainder of the requests in either the first set or the second set of requests is serviced with a response from the stateless computing platform, the stateful server is further configured to:

identify whether the remainder of the requests in one of the first set and the second set of requests is serviced before other one of the first set and the second set of requests by the stateless computing platform; and

responsive to identifying that the remainder of the requests in one of the first set and the second set of requests is serviced before other one of the first set and the second set of requests, cancel the remainder of the requests in the other one of the first set and the second set of requests issued to the stateless computing platform.

15. The system of claim 11 , wherein the stateful server is further configured to:

determine whether servicing of the cumulative percentage of requests in the first set of requests by the stateless computing platform to meet the threshold is possible; and

responsive to determining that the servicing of the cumulative percentage of requests in the first set of requests by the stateless computing platform to meet the threshold is impossible, reissue at least a portion of the first set of requests to the stateless computing platform based on a tradeoff between a cost of reissue and servicing of the first set of requests.

16. The system of claim 11 , wherein the stateful server is further configured to:

determine a failure rate of the stateless computing platform; and

determine the threshold based on the failure rate of the stateless computing platform.

17. The system of claim 16 , wherein the threshold is a difference between 100 and a multiple of the failure rate.

18. The system of claim 16 , wherein a number of times that the second set of requests is reissued to the stateless computing platform is based on the failure rate.

19. The system of claim 16 , wherein the failure rate is one of a statistical failure rate and a dynamically measured failure rate.

20. The system of claim 11 , wherein the threshold is set in a range from 85 percent to 99 percent.

Assignments (2)

Continuity (2)

References Cited (16)