(PT)-T-3, a Parameter-based Performance Prediction Tool, is an interactive performance estimator that helps users tune scientific Fortran programs. It detects bottlenecks in the program, identifies the causes of perfo...
详细信息
(PT)-T-3, a Parameter-based Performance Prediction Tool, is an interactive performance estimator that helps users tune scientific Fortran programs. It detects bottlenecks in the program, identifies the causes of performance problems, and guides users in selecting effective program transformations to gain performance. Four of the most critical aspects of parallel programs are automatically estimated: load balance, data locality, communication, and computation overhead. Operating as an integrated tool of the Vienna Fortran Compilation System, P3T is enabled to aggressively exploit considerable knowledge about the compiler's analysis information and code restructuring strategies. The author evaluates existing features and describes substantial enhancements in three key areas: the graphical user interface, performance parameters, and input programs. The tool's graphical user interface directs the user to bottlenecks in a computation that prevent the program from performing well. In addition, it allows performance data to be filtered and visualized at various levels of detail. The author discusses a new performance parameter that models computation time. Experiments show how (PT)-T-3 can successfully guide the user in performance tuning of parallel programs. The results demonstrate the feasibility of automatic performance prediction for parallel programs.
This paper describes the implementation of transmission-line matrix (TLM) method algorithms on a massively parallel computer (DECmpp 12000), the technique of distributed computing in the UNIX environment, and the comb...
详细信息
This paper describes the implementation of transmission-line matrix (TLM) method algorithms on a massively parallel computer (DECmpp 12000), the technique of distributed computing in the UNIX environment, and the combination of TLM analysis with Prony's method as well as with autoregressive moving average (ARMA) digital signal processing for electromagnetic field modelling. By combining these advanced computation techniques, typical electromagnetic field modelling of microwave structures by TLM analysis can be accelerated by a few orders of magnitude.
暂无评论