When parallel applications do not fully utilize the cores that are available to them they are missing the opportunity to have better performance. Sometimes threads have to wait for other threads. I call the code segments that make other threads wait bottlenecks. Examples of these bottlenecks include contended critical sections, threads arriving late to barriers and the slowest stage of a pipelined program. Other times all threads are running but some of them, which I call lagging threads, are making less progress, setting the stage to become bottlenecks. My thesis proposes identifying the code segments that are more critical for performance and efficiently accelerating them using faster cores, by either migrating execution to large cores of an Asymmetric Chip Multi-Processor (ACMP) or executing locally on DVFS-accelerated cores. The key contribution of this dissertation is a Utility of Acceleration metric that combines a measure of the acceleration for each code segment with a measure of its criticality. This metric enables meaningful comparisons to decide which bottlenecks or lagging threads to accelerate with each of the available acceleration mechanisms. My evaluation shows significant performance improvement for single multithreaded applications and sets of multiple single- and multi-threaded applications, and also reduction in energy-delay product due to the efficient utilization of the available acceleration mechanisms. / text

Communicating Sequential Processes (CSP) is a formal language whose primary purpose is to model and verify concurrent systems. The CSP++ toolset was created to embody the concept of selective formalism by making machine-readable CSPm specifications both executable (through the automatic synthesis of C++ source) and extensible (by allowing the integration of C++ user-coded functions). However, these user-coded functions were limited by their inability to share data with each other, which meant that their application was constrained to solving simple problems in isolation. We extend CSP++ by providing user-coded functions in the same CSP process with safe access to a shared storage area, similar in concept and API to Pthreads' thread-local storage, enabling cooperation between them and granting them the ability to undertake more complex tasks without breaking the formalism of the underlying specification. This feature's utility is demonstrated in our line-following robot case study.