EMME is a research project funded by the Hasler Foundation and it is based on previous work carried out during the MULTICUBE project. 

Programmable multi-core and many-core platforms increase exponentially the challenge of task mapping and scheduling, provided that enough task-parallelism does exist for each application. When considering that multiple applications are executed concurrently on the same multi-core platform and are competing to access system resources, a Run-time Resource Management (RRM) layer should be integrated in the OS in order to arbiter about resource distribution. The RRM should take decisions in order to maximize platform performance while minimizing nonfunctional costs such as energy or power consumption. This problem is intractable with a pure run-time approach since too many operating configurations should be considered. 

With the present project, we plan to provide a methodology for efficient Run-time Resource Management suitable for future MPSoC platforms. The work will address the analysis of efficient run-time heuristics able to identify near-optimal operating configurations without incurring in long run-time overheads. 

The outcome of this project will be an open source prototype framework for the analysis and evaluation of different RRM techniques given a target computing platform.