zbMATH — the first resource for mathematics

On using SPiDER to examine and debug real-world data-parallel applications. (English) Zbl 0997.68618
Malyshkin, Victor (ed.), Parallel computing technologies. 6th international conference, PaCT 2001, Novosibirsk, Russia, September 3-7, 2001. Proceedings. Berlin: Springer. Lect. Notes Comput. Sci. 2127, 211-225 (2001).
Summary: Debuggers are used to control the state of many processes, to present distributed information in a concise and clear way, to observe the execution behavior, to detect and to locate programming errors. In this paper we briefly describe the design of SPiDER which is an interactive source-level debugging system for both regular and irregular High Performance Fortran programs. SPiDER allows to inspect a single process of a parallel program or to examine the entire program from a global point of view. A sophisticated visualization system has been developed and included in SPiDER to visualize data distributions, data-to-processor mapping relationships, and array values. SPiDER enables a programmer to dynamically change data distributions as well as array values. For arrays whose distribution can change during program execution, an animated replay displays the distribution sequence together with the associated source code location. Array values can be stored at individual execution points and compared against each other to examine execution behavior (e.g. convergence behavior of a numerical algorithm). SPiDER has been fully implemented and is currently being used for the development of various real-world applications. Several experiments will be presented that demonstrate the usefulness of SPiDER.
For the entire collection see [Zbl 0969.00083].
68U99 Computing methodologies and applications
68N19 Other programming paradigms (object-oriented, sequential, concurrent, automatic, etc.)
Full Text: Link