Visualization and performance prediction of multithreaded Solaris programs by tracing kernel threads

Performance Optimization Using Extended Critical Path Analysis in Multithreaded Programs on Multiprocessors

“…These methods have been implemented in a performance optimization tool called VPPB [1,2]. The tool pinpoints what parts of the code to optimize without the need of a multiprocessor.…”

“…The extended critical path analysis has been implemented in a tool called VPPB [1,2]. The extended critical path analysis increases the capacity of the tool in a natural way.…”

“…Thus, the behavior of the program is traced. Additional information about the preemptive scheduling of the LWPs (lightweight processes) [16] is also collected with a Solaris system command called prex described in [1]. The next step is to simulate the recorded information.…”

“…We are then able to probe each invocation of each function. We use the same kind of recording probes, TNF (trace normal form), as previously used in the tool [1]. To simplify the implementation of the parser it assumes that the compiler uses the default optimization.…”

QoE-based Cross-Layer Design of Mobile Video Systems: Challenges and Concepts

“…[15], [16]. There have also been efforts towards automatic parallelization and efforts for providing software development tools for writing parallel terminal/client software [17]. The problem of finding real-time scheduling algorithms with guaranteed performance characteristics for multiprocessors and singleprocessors has been studied extensively and a variety of promising results exist today, both for periodic and aperiodic cases, e.g.…”