Hybrid Heuristic for Scheduling Data Analytics Workflow Applications in Hybrid Cloud Environment

Effective scheduling is a key concern for the execution of performance driven applications, such as workflows in dynamic and cost driven environment including Cloud. The majority of existing scheduling techniques are based on meta-heuristics that produce good schedules with advance reservation given the current state of Cloud services or heuristics that are dynamic in nature, and map the workflow tasks to services on-the-fly, but lack the ability of generating schedules considering workflow-level optimization and user QoS constraints. In this paper, we propose an Adaptive Hybrid Heuristic for user constrained data-analytics workflow scheduling in hybrid Cloud environment by integrating the dynamic nature of heuristic based approaches as well as workflow-level optimization capability of meta-heuristic based approaches. The effectiveness of the proposed approach is illustrated by a comprehensive case study with comparison to existing techniques.