Remotely managing execution of jobs in a cluster computing framework

A method, comprising:

• receiving, by a master device and from a managing device, a request for an execution of a plurality of jobs;
• providing, by the master device and to a driver device, the plurality of jobs and execution criteria for the execution of at least one of the plurality of jobs;
• receiving, by the master device and from the driver device, failure information associated with the execution of the at least one of the plurality of jobs according to the execution criteria;
• receiving, by the master device, from the managing device, and based on receiving the failure information, a request to retry the execution of the at least one of the plurality of jobs;
• and
• sending, by the master device and based on receiving the request to retry the execution of the at least one of the plurality of jobs, the at least one of the plurality of jobs, and either the execution criteria or modified execution criteria, to another master device for another execution of the at least one of the plurality of jobs according to the execution criteria or the modified execution criteria, wherein the other master device is associated with another driver device.

A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:

• one or more instructions that, when executed by one or more processors of a master device, cause the master device to:
• receive a request for an execution of a plurality of jobs;
• provide, to a driver device, the plurality of jobs and execution criteria for the execution of at least one of the plurality of jobs;
• receive, from the driver device, failure information associated with the execution of the at least one of the plurality of jobs according to the execution criteria;
• and
• send, based on receiving the failure information, the at least one of the plurality of jobs, and either the execution criteria or modified execution criteria, to another master device for another execution of the at least one of the plurality of jobs according to the execution criteria or the modified execution criteria, wherein the other master device is associated with another driver device.