Virginia Tech - Wake Forest University School of Biomedical Engineering & Sciences
Computer Science Technical Reports @Virginia TechNot a member yet
997 research outputs found
Sort by
Implementation of an Enhanced Distributed Object Model using Actors
No full textMost common object models of distributed object systems have a limited set of object-oriented features, lacking the advanced features of 'polymorphism' (an abstraction mechanism that represents a quality or state of being able to assume different forms) and 'concurrency' (the ability to have more than one thread of execution in an object simultaneously). The lack of support for advanced features is a serious limitation because it restricts the development of new components and limits reuse of existing components that use these advanced features. In this paper, the Interoperable Common Object Model (ICOM) centered on statically typed object-oriented languages is presented. The ICOM model is an attempt to elevate common object models (with the advanced features of polymorphism and concurrency) closer to the object models of statically typed object-oriented languages. Specific features of the ICOM model include: remote inheritance, method overloading, parameterized types, and guard methods. This paper focuses on how the actor model and reflection techniques are used to develop a uniform implementation framework for the ICOM object model in C++ and Modula-3. Two key features of the ICOM objects are discussed, remote inheritance and atomicity, along with the distributed compilation architecture of the ICOM framework emphasizing the role of the actor model in creating a simple underlying framework is emphasized
Supporting Worker Independence in Collaboration Transparency
No full textConventional collaboration-transparency systems are inefficient in their use of network resources and lack support for key groupware principles: concurrent work, relaxed WYSIWIS, and group awareness. We present an alternative implementation approach to collaboration transparency that provides many features previously seen only in collaboration-aware applications. Our approach is based on an object-oriented replicated architecture where selected single-user interface objects are dynamically replaced by multi-user extensions. The replacement is transparent to the single-user application and its developer. As an instance of this approach, we describe its incorporation into a new Java-based collaboration-transparency system, called Flexible JAMM. We conducted an empirical study to evaluate the effectiveness of Flexible JAMM versus a representative conventional collaboration-transparency system, NetMeeting. Completion times were significantly faster in a loosely-coupled task using Flexible JAMM, and were not adversely affected in a tightly-coupled task. Accuracy was unaffected by the system used. Participants greatly preferred Flexible JAMM. The evaluation validates our aim of supporting multiple styles of collaboration
Analysis of Bottlenecks in International Internet Links
No full textWe report on preliminary results of analysis into the sources of congestion in international Internet routes from the United States to several countries: Brazil and South Africa. Using the pathchar and traceroute tools, we identified which links in a variety of routes are a bottleneck. The measures used were throughput, round trip delay, and router queueing delays. Measurements with ping were used to validate the round trip times obtained with pathchar
Improving Concurrency in Common Object Models
No full textMost common object models of distributed object systems lack features related to 'concurrency' (the ability to have more than one thread of execution in an object simultaneously). The lack of support for concurrency restricts the development of new components and limits reuse of existing components that use these advanced features. In this paper, the concurrency features of the Interoperable Common Object Model (ICOM) centered on statically typed object-oriented languages, is presented. The ICOM model is an attempt to elevate common object models (with the advanced features of concurrency) closer to the object models of statically typed object-oriented languages. Specific features of the ICOM object model include: atomic objects and guard methods. The actor model is used to develop a uniform implementation framework for the ICOM object model in C++ and Modula-3
Flexible Collaboration Transparency
No full textThis paper presents two distinct contributions: First, we present a critique of collaboration transparency as it is currently implemented in contrast to collaboration-aware implementations. We find conventional collaboration-transparency systems lacking in terms of efficient use of network resources and support for key groupware principles: concurrent work, relaxed WYSIWIS, and group awareness. Second, we examine the causes of these deficiencies, and then present an alternative implementation approach based on an object-oriented replicated architecture where selected single-user interface objects are dynamically replaced by multi-user extensions. The replacement is transparent to the single-user application and its developer. As an instance of this approach, we described its incorporation into a Java-based collaboration-transparency system, called JAMM
Collaboration Transparency in Java through Event Broadcasting
No full textWidespread use of the Internet for education and research is yielding many opportunities for network-based synchronous collaboration. For example, the Java runtime environment provides a platform independent vehicle for new collaborative applications. While many toolkits are becoming available that support development of collaborative applications, they do not enable collaborative use of existing single-user Java applets, a process called collaboration transparency. This paper discusses two approaches to collaboration transparency: display broadcasting and event broadcasting. We then consider the suitability of each within the context of the Java runtime environment. Unfortunately, simple modifications to the standard Java class libraries as they are currently implemented are not sufficient to support collaboration transparency. We describe the problems and suggest solutions
Parallel Adaptive GMRES Implementations for Homotopy Methods
No full textThe success of homotopy methods in solving large-scale optimization problems and nonlinear systems of equations depends heavily on the solution of large sparse nonsymmetric linear systems on parallel architectures. Iterative solution techniques, such as GMRES(k), favor parallel implementations. However, their straightforward parallelization usually leads to a poor parallel performance because of global communication incurred by processors. One variation of GMRES(k) considered here is to adapt the restart value k for any given problem and use Householder reflections in the orthogonalization phase to achieve high accuracy and reduce the communication overhead
Multidisciplinary Optimization of a Supersonic Transport Using Design of Experiments Theory and Response Surface Modeling
No full textThe presence of numerical noise in engineering design optimization problems inhibits the use of many gradient-based optimization methods. This numerical noise may result in the inaccurate calculation of gradients which in turn slows or prevents convergence during optimization, or it may promote convergence to spurious local optima. The problems created by numerical noise are particularly acute in aircraft design applications where a single aerodynamic or structural analysis of a realistic aircraft configuration may require tens of CPU hours on a supercomputer. The computational expenses of the analyses coupled with the convergence difficulties created by numerical noise are significant obstacles to performing aircraft multidisciplinary design optimization. To address these issues, a procedure has been developed to create noise-free algebraic models of subsonic and supersonic aerodynamic performance qualities for use in the optimization of high-speed civil transport (HSCT) aircraft configurations. This procedure employs methods from statistical design of experiments theory and response surface modeling to create the noise-free algebraic models. Results from a sample HSCT design problem involving ten variables are presented to demonstrate the utility of this method
Convergence of Trust Region Augmented Lagrangian Methods Using Variable Fidelity Approximation Data
No full textTo date the primary focus of most constrained approximate optimization strategies is that application of the method should lead to improved designs. Few researchers have focused on the development of constrained approximate optimization strategies that are assured of converging to a Karush-Kuhn-Tucker (KKT) point for the problem. Recent work by the authors based on a trust region model management strategy has shown promise in managing the convergence of constrained approximate optimization in application to a suite of single level optimization test problems. Using a trust-region model management strategy, coupled with an augmented Lagrangian approach for constrained approximate optimization, the authors have shown in application studies that the approximate optimization process converges to a KKT point for the problem. The approximate optimization strategy sequentially builds a cumulative response surface approximation of the augmented Lagrangian which is then optimized subject to a trust region constraint. In this research the authors develop a formal proof of convergence for the response surface approximation based optimization algorithm. Previous application studies were conducted on single level optimization problems for which response surface approximations were developed using conventional statistical response sampling techniques such as central composite design to query a high fidelity model over the design space. In this research the authors extend the scope of application studies to include the class of multidisciplinary design optimization (MDO) test problems. More importantly the authors show that response surface approximations constructed from variable fidelity data generated during concurrent subspace optimizations (CSSOs) can be effectively managed by the trust region model management strategy. Results for two multidisciplinary test problems are presented in which convergence to a KKT point is observed. The formal proof of convergence and the successfull MDO application of the algorithm using variable fidelity data generated by CSSO are original contributions to the growing body of research in MDO
Professionalism in Computing: A Web-Based Learning System
No full textStarting from a project to develop a digital library for a Computer Science course studying social impact and computer ethics, a highly interactive Web-based learning system is in the process of development. Capable of supporting a variety of teaching/learning environments, the system is intended to support teachers whose disciplinary primary interest is other than social impact and ethics