Back to results
US 12,561,121 B2Granted
Methods and system for performing checkpointing for fault tolerance in application
AssigneeMoreh Corp.
InventorsGangwon Jo, Jungho Park
Life of this patent
Estimating the term…
Abstract
A method for performing checkpointing for fault tolerance of an application is provided, which is performed by one or more processors, and includes storing accelerator data associated with an application at a specific time point as a checkpoint, storing operation data associated with the application performed after the specific time point, and performing application fault tolerance based on the checkpoint and the stored operation data.
Claims20
A method performed by one or more computing devices, the method comprising:
- storing, at a first time point, at least one checkpoint of application data of an application, wherein a first subset of the application data corresponds to write access;
- executing, based on an operation execution message associated with the application data, an operation associated with the application data by:
- based on a first completion status of a first vector of the first subset of the application data, suspending execution of an operation associated with the first subset of the application data;
- and
- based on a second completion status of a second vector of a second subset of the application data, executing an operation associated with the second subset of the application data, wherein the operation associated with the first subset of the application data and the operation associated with the second subset of the application data are executable in parallel;
- storing, after the first time point, operation data that indicates one or more operations performed by the application after the first time point;
- and
- performing, based on one of the at least one checkpoint and the stored operation data, application fault tolerance.
5 more elements
An information processing system, comprising:
- a communication device;
- one or more processors;
- and
- memory storing instructions that, when executed by the one or more processors, cause the information processing system to:
- store, at a first time point, at least one checkpoint of application data of an application, wherein a first subset of the application data corresponds to write access;
- execute, based on an operation execution message associated with the application data, an operation associated with the application data by:
- based on a first completion status of a first vector of the first subset of the application data, suspending execution of an operation associated with the first subset of the application data;
- and
- based on a second completion status of a second vector of a second subset of the application data, executing an operation associated with the second subset of the application data, wherein the operation associated with the first subset of the application data and the operation associated with the second subset of the application data are executable in parallel;
- storing, after the first time point, operation data that indicates one or more operations performed by the application after the first time point;
- and
- performing, based on one of the at least one checkpoint and the stored operation data, application fault tolerance.
9 more elements
Dependencies are inferred from the claim text.