Visualization is an effective tool for understanding the execution of complex programs. Over the course of a visualization session a user may employ a wide range of graphical representations. Adaptable visualization displays are needed to meet these diverse interests and applications. This paper presents an approach that supports interactive creation and refinement of visualizations via on-screen manipulation of existing visual representations of the program. By acting upon visualizations on the screen, users control what aspects of the computation and what properties are displayed. An action modifies the visualization in a given program state, but users need to have a means of defining how the visualization reacts to changes in the program state. We present a manner of specifying the effect of actions on a visualization for future program states. To achieve their effect and update the program representation, actions are automatically re-applied, in the order they were performed, for each state of the computation. The interpretation of user action is based on a model for reasoning about visualizations that focuses on what information is visually communicated and is independent of the exact choice of graphical elements. We define a set of operations on the model that allow users to take advantage on the information on the screen and explore related parts of the computation.

This paper gives an overview of our work in determining when mobile agents are appropriate for monitoring distributed applications. The high degree of variability resulting from the interaction between the users, applications, and the computing environment makes succinct descriptions of when mobile agents are appropriate very dicult. Our approach has been to identify the basic classes of variables in the evaluation formula and then to perform empirical tests to identify which aspects of the various components has the greatest impact. These results will then be used to identify guidelines as to when mobile agents are advantageous for the monitoring of distributed systems.

This paper presents the design, prototype implementation and experimental evaluation of DIOS++, an infrastructure for enabling rule-based management and control of distributed scientific applications. DIOS++ provides: (1) abstractions to enhance existing application objects with sensors and actuators for runtime interrogation and control, access policies to control accesses to sensors/actuators and rule interfaces, and rule agents to enable autonomic monitoring and steering, (2) a hierarchical control network that connects and manages the distributed sensors and actuators, enables external discovery, interrogation, monitoring and manipulation of these objects at runtime, and facilitates dynamical and secure definition, modification, deletion and execution of rules for autonomic application management and control. The framework is currently being used to enable autonomic monitoring and control of a wide range of scientific applications including oil reservoir, compressible turbulence and numerical relativity simulations.

Large and complex information requires arbitrarily extensive displays for an exact, comprehensive visualization. This paper introduces a technique, time-multiplexing, for presenting such information that concentrates on the efficient management of time for a typical display. Time multiplexing complements classical approaches, which are graphics oriented and make use of spatial and perceptual features of the display to convey a view of the data at various levels of complexity. Time multiplexing offers a live virtual tour of the important aspects of the information by continuously iterating though a set of "interesting" visualizations. The interestingness of the views is periodically approximated to ensure that they conform to the user's shifting intentions and to the underlying changing information. Techniques to correlate different presentations timelines to improve the effectiveness of time-multiplexing visualization are discussed.

One of the greatest challenges facing the software community today is the increasing complexity of software. Complexity limits understanding, making it difficult to evaluate the correctness, reliability, and performance of a system. Coupled with visualization, monitoring can provide users with insight into an application's behavior. Monitoring can also be used in conjunction with automated tools to adaptively tune performance.