Exploiting Linux Control Groups for Effective Run-Time Resource Management
Source: Politecnico di Bari
The extremely high technology process reached by the silicon manufacturing (under the 32nm) has led to production of computational platforms and SoC, featuring a considerable amount of resources. Whilst from one side such multi- and many-core platforms show growing performance capabilities, from the other side they are more and more affected by power, thermal and reliability issues, as well as target of congested workload usage scenarios. Effective usage of the resources should take into account both the application requirements and resources availability, with an arbiter, namely a Resource Manager, which is in charge to solve the resource contention among demanding applications.