One of the more bothersome aspects of developing a parallel program is that of monitoring the behavior of the program for debugging and performance tun-ing. Often there is no intrinsic support for examining program state and dynamics in a parallel program-ming system, let alone a useful interface for analyz-ing or visualizing that information. This paper dis-cusses an enhanced tracing facility and tracing tool for PVM (Parallel Virtual Machine), a message passing library for parallel processing in a heterogeneous en-vironment. PVM supports mixed collections of work-station clusters, shared-memory multiprocessors, and MPPs. The upcoming release of PVM, Version 3.4, contains a new and improved tracing facility which provides more flexible and efficient access to run-time program information. This new tracing system sup-ports a buffering mechanism to reduce the perturba-tion of user applications caused by tracing, and a more flexible trace event definition scheme which is based on a self-defining data format. The new scheme ex-pedites the collection of program execution histories, and allows for integration of user-defined custom trace events. The tracing instrumentation is built into the PVM library, to avoid re-compilation when tracing is desired, and supports on-the-fly adjustments to each task’s trace event mask, for control over the level of tracing detail. Along with this new tracing facility, the graphical console and monitor XPVM has been up-dated to provide better access to the new tracing func-tionality. Several new views have been implemented to utilize the additional tracing information now possi-ble, including user-defined events. The XPVM system has also been optimized to provide better real-time monitoring capabilities. *Research supported by the Applied Mathematical Sciences

Software process is how an organization goes about developing or maintaining a software system. It is the methodology employed when people use machines, tools, and artifacts to create a product. Recent work has applied formal modeling to software process, with the hope of reaping the benefits of unambiguous and analyzable formalisms. Yet industry has been slow to adopt formal model technologies. Two reasons are that it is costly to develop a formal model and, once developed, there are no methods to ensure that the model indeed reflects reality. This thesis develops techniques for process event data analysis that help solve these two problems, which are termed process discovery and process validation. For process discovery, event data captured from an on-going process is used to generate a formal model of process behavior. To do this, results from the field of grammar inference are applied, and a new method is also developed. The methods are shown to be efficient and practical to use in...

XPVM is a graphical console and monitor for PVM. XPVM provides a graphical interface to the PVM console commands, along with animated views to monitor the execution of PVM programs. These views provide information about the interactions among tasks in a parallel PVM program to assist in debugging and performance tuning. This report serves as a User's Guide for XPVM 1.0, the first released version of the interface, including all patch levels 1.*.*. - v - 1. Introduction Often when developing a parallel program it is difficult to determine the program's behavior due to the many concurrently executing threads of control. However, it is useful for the programmer to evaluate whether the program is running as expected or performing its work efficiently. One useful approach to this problem is the use of program visualization [3] to visually depict the program's execution using computer graphics. XPVM is a graphical console and monitor for PVM [6]. Using the X Window System, it provides a ...

In developing distributed applications and services there is a need to be able to set up and run tests on a set of processes. The experiment might be to obtain performance data, to test the processes' behaviour, or to evaluate an application management strategy. Common requirements are ffl to load and run special versions of at least some of the software, often on multiple nodes of a network, ffl to initialize the software in a wellcontrolled way, so the tests may be repeatable, ffl to monitor execution and collect results for post-execution or real-time analysis. DECALS 1 is a controlling framework to support the testing of distributed applications consisting of many processes on a network. There may be any number of experiments; in each experiment a set of processes is loaded on specified workstations, and each process is initialized with data for the particular experiment. The configuration of the experiment, and the data state of the processes, can be controlled down to any l...

This paper considers the design of a reusable software template for a parallel data-farm which uses demand-based load-balancing. A feature of the farm is integral instrumentation. A design example is given for a hybrid processor message-passing machine (the Paramid) in which monitoring is accomplished by an instrumented interface program. Other aspects of the design are use of buffering to mask communication latency, an asynchronous multicast provision, and a controlled interface to the worker functions. Trace material is discussed from two examples when the template design was used to monitor real-time, continuous-flow applications.

This paper describes a methodology that provides detailed predictive performance information throughout the software design and implementation cycles. It is structured around a hierarchy of performance models that describe the computing system in terms of its software, parallelization, and hardware components.The methodology is illustrated with an initial implementation, the performance analysis and characterization environment (PACE) system, which provides information concerning execution time, scalability, and resource use. A principal aim of the work is to provide a capability for rapid calculation of relevant performance numbers without sacrificing accuracy

ion (EBBA). Some of these methods will be briefly introduced here. The simplest way of analyzing runtime information consists of collecting statistics from the individual event-records. ParaGraph [HE91b] and the Crystal [RR89b] are typical examples of tools that use this approach. Among the gathered statistics are cumulative busy/idle times, cumulative communication times, number of bytes sent, etc. More complex analysis is provided by the integration into the monitoring environment of statistical analysis packages, that allow the interactive analysis of the trace-data. The SIMPLE environment adopted this approach by integrating the data analysis and graphics package S from AT&T [Moh90]. The Event Based Behavioral Abstraction (EBBA) approach is more than simply a way of analyzing event-records, and constitutes a complete high-level approach to debugging [Bat89]. Globally speaking the approach consists of constructing high-level models that describe the expected behavior of the monit...

ion: program execution ! program data representation (1) Visualisation: program data representation ! pictorial representation (2) Here, abstraction takes the form of instrumentation and data collection, while visualisation comprises the last two steps of Figure 6. Abstraction merely represents the types of information or program entities which are essential in driving the final visualisation. For example, in visualising a parallel program, it can be useful to derive information on the utilisation of processors. On the other hand, visualisation is a step used to present the program entities or information in final displays. For example, a single stream of information on processor utilisation can be represented as a processor-usage profile, such as the Kiviat diagram (see Figure 3), and as a concurrency profile that shows the number or percentage of processors executing concurrently at each point in time 11 . Hardware Architecture Uniprocessor Multiprocessor Distributed Parallel Op...

Researchers and practitioners in the area of parallel and distributed computing have been lacking a portable, flexible and robust distributed instrumentation system. We present the Baseline Reduced Instrumentation System Kernel (BRISK) that we have developed as a part of a realtime system instrumentation and performance visualization project. The design is based on a simple distributed instrumentation system model for flexibility and extensibility. The basic implementation poses minimalistic system requirements and achieves high performance. We show evaluations of BRISK using two distinct configurations: one emphasizes isolated simple performance metrics; and the other, BRISK's operation on distributed applications, its built-in clock synchronization and dynamic on-line sorting algorithms. 1 Introduction and Related Work Designers and users of parallel and distributed systems have applied a variety of monitoring methods and instrumentation techniques to gather information for testing...

PG RT is an environment for the integration of tools and systems for instrumentation, performance visualization, and analysis of complex real-time systems. A part of PG RT is the Visual Objects framework for rapid prototyping and development of object-oriented, online, adaptive, application-specific performance visualizations. PG RT uses CORBA technology to provide real-time applications and systems with convenient access to visual objects. Many complex visual objects may be distributed over a number of graphical workstations in a local area network to provide rich visualization in real time. The visual objects can be directly accessed from a distributed real-time application, or--- in order to eliminate the intrusion of CORBA to the application---via a distributed instrumentation system called BRISK, which is a component of PG RT . We demonstrate the PG RT environment by instrumenting and visualizing a distributed real-time multimedia application. 1 Introduction An abili...