Job Batch Scheduling in Workflow-as-a-Service Platforms

In this work, we propose an approach for executing science-intensive applications within the framework of the new Workflow-as-a-Service (WaaS) concept. WaaS platforms, the so-called multitenant environments, provide efficient mechanisms for managing continuous and heterogeneous job flows in cloud computing. The workflow execution schedule is built using the critical jobs method (CJM), which allows the scheduling of information-dependent tasks within the directed acyclic graph (DAG) model. Nevertheless, it must be adjusted to consider the actual dynamics of the utilization of resources during each scheduling cycle. In this scenario, along with the complexity of scheduling composite workflows, there is an additional issue: the efficient management of cloud resources, namely, determining the start and shutdown time of virtual machines (VMs) according to the available economic policy. To handle this problem, we suggest a modification to CJM. The idea behind this modification is that the collision resolution stage is implemented in an independent module, which assigns workflow tasks to specific instances of resources provided by the cloud infrastructure provider. The resulting solution combines several heuristics to optimize cloud resource management for WaaS platforms. Experiments with real-world workflows prove the optimization efficiency of the suggested approach.