In today’s world, data communication and execution- whether online or offline- became essential in all business fields. Considering the case of offline execution, data files are collected in groups called “input batches” and processed using available resources to get the required aggregated output in the form of “output batch”, this is called “Data Batching Process”. In most batch process applications, the target is to execute required tasks within specified time frame while fulfilling all predecessors requirements and constraints set by the client using minimum resources to achieve this goal. The data batch process terms and conditions are stated in the contract between service provider and the client - Known as Service Level Agreement (SLA)-.

During research it was noted that few researchers tried including cost in their study. In this paper, the attempt is to consider the aforementioned resources, predecessors and constraint factors while trying to minimize the data batch process cost which include: servers and software basic leasing cost, additional resources rental cost needed in case of overload, penalty cost of failing to execute the batch process as per the SLA and the opportunity cost of idling a resource for any period of time due to inefficient task allocation.

Keywords—Data batch process, scheduling, cost, parallel processors, process time.