Resiliency control engine for network service mesh systems
AssigneeBank of America Corporation
InventorSundarrajan Raman
Life of this patent
Estimating the term…
Abstract
A resiliency controller may monitor execution of a plurality of services, each service of the plurality of services associated with a different container of a containerized computing system. In response to requests received from a plurality of connected client computing systems, the resiliency controller predicts, based on historical container operation information, a first predicted latency associated with the first request and a second predicted latency associated with the second request. The resiliency controller associates a first latency response to the first request and a second latency response to the second request, and initiates a selected latency response based on an indication that at least one of the first predicted latency or the second predicted latency has been exceeded.
Claims20
Non-transitory computer readable media storing instructions that, when executed by at least one processor, cause a containerized computing system to:
- receive, via a network, a request message from a first client computing system of a plurality of computing systems;
- predict, based on a message priority of the request message and system latency information, a predicted cause of latency associated with execution of the request message by a target service and an associated latency response action;
- determine, by a resiliency control engine processing a regressor algorithm and a weighted combination of inputs, a latency control action to be performed in response to a latency condition, wherein the inputs comprise the predicted cause of latency, a client priority parameter, and an availability parameter;
- and
- initiate execution of the request message by the target service.
A method comprising:
- monitoring execution of a plurality of services associated with a different container of a plurality of containers of a containerized computing system;
- predicting, based on historical container operation information, a first predicted cause of latency associated with a first request received from a first client computing system and a second predicted cause of latency associated with a second request received from a second client computing system;
- determining, by a resiliency control engine processing a regressor algorithm based on a weighted combination of the first predicted cause of latency and the second predicted cause of latency, a first latency control action and a second latency control action to be performed in response to a latency condition;
- and
- associating the first latency control action to the first request and the second latency control action to the second request;
- and
- initiating, by a latency control engine, a latency response based on an indication that one or both of a first predicted latency and a second predicted latency has been exceeded.
An apparatus comprising:
- at least one processor;
- and
- memory storing instructions that, when executed by the at least one processor, cause the apparatus to:
- predict, based on a message priority of a request message received from a first client computing system of a plurality of client computing systems and system latency information, a predicted cause of latency and an associated latency response action, wherein the predicted cause of latency is associated with execution of the request message by a target service;
- determine, by a resiliency control engine processing a regressor algorithm based on a weighted combination of the predicted cause of latency, a client priority parameter, and an availability parameter, a latency control action to be performed in response to a latency condition;
- and
- initiate execution of the request message by the target service.
Dependencies are inferred from the claim text.