Cedell Alexander

Learn More
Program activity graphs (PAGs) can be constructed from timestamped traces of appropriate execution events. Information about the activities on the k longest execution paths is useful in the analysis of parallel program performance. In this paper, four algorithms for finding the near–critical paths of PAGs are compared, including a best–first search (BFS)(More)
Program activity graphs can be constructed from timestamped traces of appropriate execution events. Information about the activities on the k longest execution paths is useful in the analysis of parallel program performance. In this paper, four algorithms for finding the near–critical paths of program activity graphs are presented and compared, including an(More)
In a previous issue, we described a study of translation buffer performance undertaken in conjunction with the design of a memory management unit for a new 32 bit microprocessor [Alex85b]. This work produced generalized results via trace-driven simulations. The address reference traces were obtained from typical multitasking UNIX workloads and have now been(More)
Durin9 the design of i memory menagement unK for a new 32 bit microprocessor (32 bit address & data), we were confounded by the lack of general & credible trenslaUon buffer (TB) performance evaluations. That led Lo this study, which we hope can benefit other designers. Although our evaluaUon is from s 32 bit, VLSI, and UNIX" perspective, mosL results r e(More)
A new multicomputer performance monitoring system is described in this paper. Where possible, the system employs portable performance monitoring instrumentation technology and leverages previous work. Trace event acquisition is hardware assisted and based on the MultiKron, a single–chip measurement solution developed at the National Institute for Standards(More)
  • 1