LEVERAGING OFFLOAD PROGRAMMING MODEL FOR LOCAL CHECKPOINTS

摘要

PROBLEM TO BE SOLVED: To provide methods, apparatus and systems for leveraging an offload programming model for local checkpoints.SOLUTION: In a computing environment implemented as one or more sources and a large number of sinks, code sections designated for offload in a job code whose execution is managed by a source are offloaded by the source to the sinks by creating offload context information. The offloaded code sections are executed by the sinks, which return result data to the source. The health of the sinks is monitored and, upon a failure, the source retrieves an offload object corresponding to the code section offloaded to the failed sink, regenerates the offload context information for the code section, and sends this to another sink for execution.SELECTED DRAWING: Figure 2