Krishna Kishore Dhara

Forcing behavioral subtyping through specification inheritance

A common change to object-oriented software is to add a new type of data that is a subtype of some existing type in the program. However, due to message passing, unchanged pearls of the program may now call operations of the new type. To avoid reverification of unchanged code, such operations should have specifications that are related to the specifications of the appropriate operations in their supertypes. This paper presents a specification technique that uses inheritance of specifications to force the appropriate behavior on the subtype objects. This technique is simple, requires little effort by the specifier, and avoids reverification of unchanged code. We present two notions of such behavioral subtyping, one of which is new. We show how to use these techniques to specify examples in C++.

Weak Behavioral Subtyping for Types with Mutable Objects

Abstract This paper studies the question of when one abstract data type (ADT) is a behavioral subtype of another, and proposes a model-theoretic notion of weak behavioral subtyping. Weak behavioral subtyping permits supertype abstraction to be a sound and modular reasoning principle in a language with mutation and limited forms of aliasing. The necessary restrictions on aliasing can be statically checked. Weak behavioral subtyping allows types with mutable objects to be subtypes of types with immutable objects.

Enhancing Application-Layer Multicast for P2P Conferencing

While application-layer multicast (ALM) can re- duce latency and permit deployment in networks without network-layer multicast, often, it does not leverage application- specific behavior. For example, the existing ALM-based peer-to- peer (P2P) conferencing systems route media packets based only on bandwidth and on transmission delay without considering P2P conferencing specific behavior and varying mixing capabilities of participating peers. We propose an enhanced hybrid ALM-based P2P conferencing architecture that can leverage participating peers with different capabilities. We also present a three-step algorithm to construct such a hybrid ALM-based P2P conference. We have built a prototype for our architecture. Results from our prototype show that our approach is more scalable than the existing ALM-based P2P conferencing systems.

Overview of Structured Peer-to-Peer Overlay Algorithms

This chapter provides an overview of structured Peer-to-Peer overlay algorithms. The chapter introduces basic concepts including geometries, routing algorithms, routing table maintenance, node join/leave behaviour, and bootstrapping of structured Peer-to-Peer overlay algorithms. Based on these key concepts, a number of key overlay algorithms are classified into categories and a brief over-view of these algorithms is presented. Finally, the chapter presents an ’on-a-glance’ comparison of the presented algorithms and provides an outlook on open research issues.

Virtual Worlds as a Tool for Enterprise Services

Online virtual worlds offer exciting new opportunities for enterprises or businesses. To exploit these opportunities enterprises need to find a way to integrate virtual world services with their real-world services, and to exploit virtual worlds rich collaboration environment to offer new or enhanced services. However, there are many challenges to realizing these goals, such as resolving virtual and real identifications, controlling communication sessions between real and virtual world entities, and managing resources in both worlds. In this paper, we study these issues and propose a framework for integrating virtual worlds into enterprise service work flow. We implemented a customer service in a VW contact center that is integrated with corresponding enterprise services. We also present a novel mechanism in virtual worlds that allow users to bring their own resources, along with the contact center resources, and share the resources with other customers or agents to resolve customer issues. Further, we present experiences from our prototype.

Dynamic peer-to-peer overlays for voice systems

Different devices, such as mobile phones, soft phones, or desktop phones, have varying processing power, bandwidth, and media capabilities. Heterogeneous P2P voice systems that are built based on a set of capabilities will not be suitable for devices that have different capabilities. In this paper, we present an architecture for P2P voice systems that can dynamically change P2P overlay mechanisms to better suit different device, user, and feature requirements of a P2P voice system. As a first step towards realizing the architecture, we propose a P2P-SIP based architecture that separates out P2P mechanisms from SIP. Our architecture allows dynamic P2P structural changes, limits bloating of the SIP protocol, and lays a foundation for a flexible hierarchical system

Reconsidering Social Networks for Enterprise Communication Services

Social networking applications leverage valuable information from social graphs and integrate communication capabilities to offer new services. However, currently, these applications are limited to communication capabilities such as click-to-call or click-to-conference. Once the communication session is established the social context and the communication context are not integrated in these services. For social network communication services such integration permits new and novel user services. Further, enabling those services in the enterprise context will enhance social networks as a communication and collaboration tool in enterprises. In this paper, we look at such services and argue that social network applications and enterprise social networking communication applications have some fundamental differences. We discuss new notions such as transient relationships, domain separation, user views, and several other issues that need to be addressed for enabling new social networking communication services for enterprises. We present a model that addresses these issues and describe our prototype implementation with a use case example.

Feature interactions between internet services and telecommunication services

The automated run-time detection of feature interactions in Web 2.0 communications applications is an important problem that has not been addressed to date. Such web-enabled communication services are constructed as reusable web widgets that can be composed by users in a web interface. Widgetizing communication features as web services can better serve users with highly customizable features, friendly user interfaces, and easier integration with other web services. However, it also introduces new feature interaction problems. As we show, these composite communication services combine web services and VoIP features in highly dynamic interfaces with application and service state that is typically distributed across multiple domains. In this paper, we present these new feature interactions and propose our solution. We present ten different feature interactions and organize them into a taxonomy. A general method for detecting FIs is given in FOL notation. We also present a coordinator plug-in mechanism for independent widgets to share feature information and state that can be implemented in todays browsers. We finally describe a run-time algorithm for FI detection that is suitable for this architecture and uses the FI notation presented in the paper.

Managing Distributed Feature Interactions in Enterprise SIP Application Servers

Several trends in SIP application server deployments in large scale telephony environments exacerbate the classic problem of feature interaction : use of distributed feature servers, mixing of legacy and green-field feature servers, and the co-existence of multiple third-party feature implementations provisioned in the same environment. Next-generation SIP application servers will include an application router (AR) to provide more application control over feature sequencing. As we discuss here, the AR can be augmented to incorporate feature interaction detection and resolution logic. We describe a novel design for run-time feature interaction detection and resolution in an environment of distributed feature servers using a SIP application server with application routing function, such as that defined in JSR 289. The approach is based on the algorithm of the Kolberg-Magill (K-M) method for feature interaction detection. Here we extend the notation of the algorithm to cover advanced call control services, enable the algorithm to work in topologies involving B2BUAs (Back-to-Back User Agents) and SBCs (Session Border Controllers), and test the approach with a substantial feature set of 32 features found in large enterprise solutions.

Mixed reality for enhancing business communications using virtual worlds

Online virtual worlds are attracting businesses that intend to offer new enterprise class services. Often, these services in virtual worlds are closely linked with the real enterprise resources. For a successful deployment of these services, a mixed reality model with communications that extend the virtual worlds to enterprise resources is required. In this paper, we take a look at this new class of collaborative applications by using a customer service application as an example. We discuss various issues in offering such a service, lay out the requirements, propose an architecture for mixed reality communications, and present our prototype implementation. We believe that this example service can introduce a mixed reality based communication paradigm that is applicable to a wide range of other business or enterprise services.