International Business Machines Corporation

Brian Cho

Pin Zhou

Kavita Chavda

Seshashayee Sankarshana Murthy

Mickey Iqbal

Sandeep Madhav Uttamchandani

Sandeep M Uttamachandani

A method and associated system for continuously optimizing data archive management scheduling. A flow network is modeled, which creates vertexes organized in multiple levels and creating multiple edges sequentially connecting the vertexes of the multiple levels. The multiple levels consist of N+1 levels denoted as LEVEL, LEVEL, . . . , LEVEL, wherein N is at least 2, wherein LEVELincludes a source vertex, wherein LEVELincludes at least one task vertex associated with at least one archive management task, wherein EDGEis directed from the vertex of LEVELto the vertex of LEVELfor K=1, . . . , N, wherein EDGEindicates that a first task associated with a first task vertex of LEVELis scheduled by a job scheduler, and wherein EDGEhas a first weight equal to a time unit to perform the first task.

Continuous optimization of archive management scheduling by use of integrated content-resource analytic model

A method and associated system for continuously optimizing data archive management scheduling. A flow network is modeled. The flow network represents data content, software programs, physical devices, and communication capacity of the archive management system in various levels of vertices such that an optimal path in the flow network from a task of at least one archive management task to a worker program of the archive management system represents an optimal initial schedule for the worker program to perform the task.

A system and associated method for continuously optimizing data archive management scheduling. A job scheduler receives, from an archive management system, inputs of task information, replica placement data, infrastructure topology data, and resource performance data. The job scheduler models a flow network that represents data content, software programs, physical devices, and communication capacity of the archive management system in various levels of vertices according to the received inputs. An optimal path in the modeled flow network is computed as an initial schedule, and the archive management system performs tasks according to the initial schedule. The operations of scheduled tasks are monitored and the job scheduler produces a new schedule based on feedbacks of the monitored operations and predefined heuristics.

Growing community of inventors

Savoy, IL, United States of America

Brian Cho