Dr. Alexandros Bartzas
Runtime Resource Runtime Resource Management Techniques for Many-core Architectures: The 2PARMA Approach,
Dr. Alexandros Bartzas, et al.,
Institute of Communications and Computer Systems, Athens, Greece,

Time: 10:40 - 11:00am
Location: Gold Room


The current trend in computing architectures is to replace complex superscalar architectures with meshes of small homogeneous processing units connected by an on-chip network. This trend is mostly dictated by inherent silicon technology frontiers, which are getting as closer as the process densities levels increase. The number of cores to be integrated in a single chip is expected to rapidly increase in the coming years, moving from multi-core to many-core architectures. This trend will require a global rethinking of software and hardware approaches. Multi-core architectures are nowadays prevalent in general purpose computing and in high performance computing. In addition to dual- and quad-core general purpose processors, more scalable multi-core architectures are widely adopted for high-end graphics and media processing.

Real-time applications, hard or soft, are raising the challenge of unpredictability. This is an extremely difficult problem in the context of modern, dynamic, multiprocessor platforms which, while providing potentially high performance, make the task of timing prediction extremely difficult. Also, with the growing software content in embedded systems and the diffusion of highly programmable and re-configurable platforms, software is given an unprecedented degree of control on resource utilization. Software generation and performance evaluation tools have to be made aware of the particularities of a certain memory hierarchy, or the dynamic features of the processor microarchitecture, such as to be able to both generate efficient code and accurately predict performance numbers. Existing approaches that are looking into runtime resource management still require big design time efforts. During this design time efforts profiling information is gathered and analysed in order to construct a runtime scheduler that can be lightweight. One can therefore note that there is a trade-of to be made between design time effort and runtime effort, where the approaches mentioned above favour more design time effort to require less runtime effort by the runtime resource manager.

In this paper we present a run-time resource manager (RTRM) for many-core architectures. This RTRM will offer adaptive task and data allocation as well as scheduling of the different tasks and the accesses to the data. Furthermore, the adequate power management techniques as well as the integration to the Linux OS will be developed. RTRM will take into account: i) the requirements/specifications of many-core architectures, applications and design techniques; ii) the dynamic compilation chain and OS support for resource management and iii) a design space exploration phase.