System Support for Scalable Coordination in Multi-core Platforms
Open AccessThere is growing desire to parallelize software and systems due to the increasing availability and utility of multi-core platforms. Ideally, the performance is desired to scale up linearly as core count increases, if there is enough parallelism can be explored inside the system. However, with more and more cores, the inter-core coordination among those parallelized software components becomes a performance challenge. As a result, system support for scalable coordination is necessary to enable applications to efficiently harness the parallel computation capacity of multi-core platforms with a wide spectrum, from embedded devices to edge cloud to rack-scale systems in data-center.This thesis investigates scalable coordination support for three different domains, each with unique system limitations and performance requirements.RT-SMR, implementation and analysis of scalable memory reclamation which provides efficient and predictable resource sharing in real-time systems;BI, a consistency model which enables data-structures sharing in non-cache-coherency rack-scale systems; andEdgeOS, a micro-kernel baaed operating system that provides fine-grained isolation for scalable, dynamic, multi-tenant Edge Clouds.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.