Back to results
US 11,240,146 B2Granted
Service request routing
AssigneeKabushiki Kaisha Toshiba
InventorTimothy David Farnham
Life of this patent
Estimating the term…
Abstract
A computer implemented method of routing service requests to service instances in a service mesh. The method comprises monitoring one or more performance characteristics of each of a plurality of network links to service instances. The method further comprises making a prediction of the network latency of transmitting a service request via each of at least two of the plurality of network links and selecting one of the plurality of service instances and routing the service request thereto based on the predictions of the network latencies.
Claims18
A method at least partially performed by electronic circuitry, the method comprising:
- monitoring at least one of a first performance characteristic of a first network link or a second performance characteristic of a second network link, the first network link being connected to a first service instance, the second network link being connected to a second service instance, the first service instance communicating with the second service instance via a service mesh;
- making a first prediction of first network latency of transmitting a service request via the first network link and a second prediction of second network latency of transmitting the service request via the second network link;
- using the first prediction of the first network latency and the second prediction of the second network latency to select one of the first service instance or the second service instance, and route the service request to the selected one of the first service instance or the second service instance;
- and
- estimating a first effective capacity of the first network link based on an observed period for which the first network link is not active and a second effective capacity of the second network link based on an observed period for which the second network link is not active.
2 more elements
A method at least partially performed by electronic circuitry, the method comprising:
- monitoring at least one of a first performance characteristic of a first network link or a second performance characteristic of a second network link, the first network link being connected to a first service instance, the second network link being connected to a second service instance, the first service instance communicating with the second service instance via a service mesh;
- making a first prediction of first network latency of transmitting a service request via the first network link and a second prediction of second network latency of transmitting the service request via the second network link;
- using the first prediction of the first network latency and the second prediction of the second network latency to select one of the first service instance or the second service instance, and route the service request to the selected one of the first service instance or the second service instance;
- and
- estimating a first effective capacity of the first network link based on an observed period for which the first network link is not active and a second effective capacity of the second network link based on an observed period for which the second network link is not active,
- wherein making the first prediction of the first network latency of transmitting the service request via the first network link comprises dividing a data size of the service request by the estimated first effective capacity of the first network link.
3 more elements
A system comprising:
- receiver circuitry configured to receive monitoring results relating to a first network link connected to a first service instance in a service mesh and a second network link connected to a second service instance in the service mesh, the monitoring results being indicative of at least one of a first performance characteristic of the first network link or a second performance characteristic of the second network link, the first service instance communicating with the second service instance via the service mesh;
- control circuitry configured to generate a first routing policy for routing a service request based on predicting a first network latency of transmitting the service request via the first network link and a second network latency of transmitting the service request via the second network link;
- and
- estimate a first effective capacity of the first network link based on an observed period for which the first network link is not active and a second effective capacity of the second network link based on an observed period for which the second network link is not active.
1 more element
Dependencies are inferred from the claim text.