Policy-based Network Management: Conflict Detection and Semantic Validation Algorithms
Researchers: Christina Zamfirescu, Hunter College
A policy system shifts the focus from configuring individual devices to managing a network in aggregate, and controlling device behavior through network policies. The system developed within this project is implemented by centralizing control functions into a single software application. At the center of such a policy system is the policy rule. Policy rules may be general and abstract or specific and concrete. In either case, policy rules represent a pairing of conditions and actions that are intended to be device and vendor independent. The composition of policy rules, along with some of the characteristics of devices that are being controlled by policy rules are also modeled. Other characteristics of devices, used to capture the semantics and relationships between different objects being managed, define how the conditions and actions represented in a Policy rule are interpreted and what effect they have on the functions of the device. This project also defines a context for the schema and semantic definitions.
Encoder Software
Researchers: K. Ravindran, City College
Transmission networks are beginning to support large volumes of data flows, thereby making congestion in network paths an unavoidable phenomenon requiring remedy by both sources and receivers. This project studied the software-level encoding and network techniques for supporting adaptive video transmissions in distributed multimedia application settings in order to adjust the sending frame rate of video sources to adapt to congestion levels in the network. Underlying this analysis is the layering of video transmissions so that the network and/or end-system elements can send or discard selective layers depending on bandwidth availability in the network. Practical applications for multi-layer video encoder development include all digital imaging industries. Beyond digital imaging, Panasonic Research has expressed interest in this study.
Collaborative Filtering Support for Locative Media: Kamida/Socialight
Researchers: Christopher Vickery, Queens College
The Persistent Java Virtual Machine (pjvm) project began as an attempt to reduce the overhead involved in the compile-test-edit cycle during the development of Java applications. The idea was to keep a single JVM “alive” across iterations of the development cycle in order to eliminate the loading and initializing of a new JVM each time. An analogy would be the use of the “++” option with IBM’s Jikes compiler, which keeps the compiler process running, with an automatic recompilation of only those source files that have changed when the user presses the Enter key. Similarly with pjvm, only those classes that have been modified would need to be reloaded into the JVM during an iteration of the development cycle. In developing the pjvm facility, we soon realized that the tool we were developing was far more useful than simply a mechanism for streamlining the development cycle. Java’s Reflection Mechanism, a key feature needed to support pjvm, provides professional and student developers with key insights about the structure and operation of code running in the JVM environment. Using the pjvm interface, the user cannot only create JVMs and load classes into them, but also instantiate classes, invoke methods, and get lists of objects, methods and classes currently residing in each JVM. Right now, pjvm is not a full-featured debugger, but with the addition of standard debugging facilities for tracing code will, we think, make pjvm the development platform of choice for serious Java developers.
Lightweight Message Queuing Protocol and Library
Researchers: Sergei Artemov, Graduate Center
The goal of this project was to construct a communication framework that would help developing mission-critical systems involving wireless networks. The suggested network protocol provides a transparent protection against message loss and/or duplication during temporal connection failures over a wireless carrier.
Identifying Web Site Semantic Structures
Researchers: Abbe Mowshowitz and Akira Kawaguchi, City College
This project aims to organize the content of a Web site based on an analysis of the semantic structure of the site’s contents. Specifically, this project will extend classification beyond localized search-based ad hoc tree structures because conventional tree representation may not capture the ways in which users search for information. Thus, information might not be found by traversing a path from the root of the classification tree. This approach will lead to improvements over simple keyword matching search capabilities for a particular site. Furthermore, the semantic structure will provide Web site designers with the means to discover similar and/or redundant information, which in turn will assist them in reconfiguring a site to achieve improved presentation and search performance.