Planning Workflow Executions over the Edge-to-Cloud Continuum

In the ever growing field of Data Science, Cloud Computing is a well established computational paradigm, while Edge Computing is an emerging and promising alternative when it comes to the novel challenges introduced in the Internet of Things (IoT) landscape. The combination of the two as a unified paradigm forms the Edge-to-Cloud continuum, which allows the execution of applications and services to span both edge and cloud resources in a transparent way. In such heterogeneous and volatile environments, the scheduling of data intensive workloads is a difficult task, usually performed manually and requiring careful and educated decisions on the type of devices used to optimally exploit the underlying hardware and achieve any user-defined higher level policy. In this paper we present the EC-Planner, a planning component for Edge-Cloud environments, which can make intelligent, automated decisions both on how and where to map arbitrary data analytics tasks to the underlying heterogeneous infrastructure, which may consist of a mix of devices and processing units, including CPUs and hardware accelerators both in the Cloud and at the Edge.