D3.5: Dynamic Partitioning Techniques

In this deliverable we focus on the description of those mechanisms that support the implementation of the kernel execution migration across dierent target devices and the collection
of runtime information metrics (such as energy consumption, memory consumption and execution time) that eventually are provided to the Dynamic Partitioning Engine (DPE) for making
decisions about where and when given kernels of the application have to be executed. The DPE is part of the partitioning framework and represents the bridge between the low level
REPARA runtime and the high-level tools where kernel scheduling decisions are made. In this deliverable, in particular, we describe the communication protocol between the DPE and the
FastFlow parallel patterns and how commands received from the DPE are executed by the pattern runtime.


Comments are closed.