PVaniM: a tool for visualization in network computing environments

Abstract

Network computing has evolved into a popular and effective mode of high performance computing. Network computing environments have fundamental differences from hardware multiprocessors, involving a different approach to measuring and characterizing performance, monitoring an application's progress and understanding program behavior. In this paper, we present the design and implementation of PVaniM, an experimental visualization environment we have developed for the PVM network computing system. PVaniM supports a two‐phase approach whereby on‐line visualization focuses on large‐grained events that are influenced by and relate to the dynamic network computing environment, and postmortem visualization provides for detailed program analysis and tuning. PVaniM's capabilities are illustrated via its use on several applications and a comparison with single‐phase visualization environments developed for network computing. Our experiences indicate that, for several classes of applications, the two‐phase visualization scheme can provide valuable insight into the behavior, efficiency and operation of distributed and parallel programs in network computing environments. © 1998 John Wiley & Sons, Ltd.

Combing the Communication Hairball: Visualizing Parallel Execution Traces using Logical Time

“…Ariadne [11], PVaniM [40], and Growing Squares [14] animated message events between processes in logical time for debugging and program comprehension purposes, but did not include physical time information needed for performance analysis. Furthermore, these visualizations portrayed at most tens of processes.…”

MOVIS: A system for visualizing distributed mobile object environments

“…The PVanim system [48] is a toolkit for creating visualizations of the execution of PVM programs. PARADE [45] in an environment for developing visualizations of parallel and distributed programs.…”

SoftArch/MTE: Generating Distributed System Test-Beds from High-Level Software Architecture Descriptions

“…The performance of deployed distributed systems has been of interest to developers for many years and has resulted in the development of a wide variety of middleware, network, database and software performance monitoring and architecture visualisation tools [27,3,8,16,34]. Many of these approaches aim to provide high-level abstractions for viewing architecture structure and/or performance results [8,16].…”

“…While Visualisations in many of these tools are often quite low-level i.e. tend to focus on low-level programmatic features rather than high-level architectural abstractions, some monitoring tools capture low-level performance information and aggregate this and present it to the user in a higherlevel abstraction [34,27]. Most however do not present performance results in a form compatible with the same visual abstractions used when modelling architecture designs making interpretation of results more difficult.…”

“…In order to enable developers to get a more accurate assessment of architectural design decisions various approaches have been developed to validate architectural designs. These include architecture-based simulation and modelling [4,24,36], performance prototypes [19,10,17], and performance monitoring and visualisation of similar, existing systems [4,34]. Simulation of architecture and middleware performance tends by its very nature to be more or less inaccurate, giving results of limited usefulness.…”