Brad Topol

Transcoding: extending e-business to new environments

The promise of e-business is coming true: both businesses and individuals are using the Web to buy products and services. Both want to extend the reach of e-business to new environments. Customers want to check accounts, access information, and make purchases with their cellular phones, pagers, and personal digital assistants (PDAs). Banks, airlines, and retailers are competing to provide the most ubiquitous, convenient service for their customers. Web applications designed to take advantage of the rich rendering capabilities of advanced desktop browsers on large displays do not generally render effectively on the small screens available on phones and PDAs. Some devices have little or no graphics capability, or they require different markup languages, such as Wireless Markup Language (WML), for text presentation. Transcoding is technology for adapting content to match constraints and preferences associated with specific environments. This paper compares and contrasts different approaches to content adaptation, including authoring different versions to accommodate different environments, using application server technology such as JavaServer pagesTM (JSPTM) to create multiple versions of dynamic applications, and dynamically transcoding information generated by a single application. For dynamic transcoding, the paper describes several different transcoding methodologies employed by the IBM WebSphereTM Transcoding Publisher product, including HyperText Markup Language (HTML) simplification, Extensible Markup Language stylesheet selection and application, HTML conversion to WML, WML deck fragmentation, and image transcoding. The paper discusses how to decide whether transcoding should be performed at the content source or in a network intermediary. It also describes a means of identifying the device and network characteristics associated with a request and using that information to decide how to transcode the response. Finally, the paper discusses the need for new networking benchmarks to characterize the server load and performance characteristics for dynamic transcoding.

Integrating visualization support into distributed computing systems

Visualization and animation tools may become extremely important aids in the understanding, verification, and performance tuning of parallel computations. Presently, however, the use of visualization has had only a limited use for enhancing parallel computation. We hypothesize that one of the primary reasons for the limited use of visualization tools in parallel program development is the difficulty of acquiring the information necessary to drive the visual display. Our approach to this impediment focuses on integrating visualization support directly into a distributed computing system. Central to this integration is the addition of a logical clock that prevents the timestamps of events from violating causality. The implementation requires the piggybacking of a negligible amount of extra header information on system messages and the impact on performance is minimal. This results in a system that produces useful visualizations with no extra effort required by the applications programmer. Also integrated into the distributed system is support which simplifies the creation of programmer-defined, application-specific visualizations, unique to each new parallel program developed.

Robust state sharing for wide area distributed applications

We present the Mocha wide area computing infrastructure we are developing. Mocha provides support for robust shared objects on heterogeneous platforms, and utilizes advanced distributed shared memory techniques for maintaining consistency of shared objects that are replicated at multiple nodes to improve performance. In addition, our system handles failures that we feel will be common in wide area environments. We have used an approach that makes use of multiple communication protocols to improve the efficiency of shared object state transfers in Mocha. We also provide an empirical evaluation of our prototype implementation for local area, wide area, and home service networks and present a sample home service application that has been programmed with the system.

Visualizing parallel simulations in network computing environments: a case study

Parallel discrete event simulation systems (PDES) are used to simulate large-scale applications such as modeling telecommunication networks, transportation grids, and battlefield scenarios. While a large amount of PDES research has focused on employing multiprocessors and multicomputers, the use of networks of workstations interconnected through Ethernet or ATM has evolved into a popular and effective platform for PDES. To improve performance in these environments, we investigate the use of graphical visualization to provide insight into performance evaluation and simulator execution. We began with a general-purpose network computing visualization system, PVaniM, and used it to investigate the execution of an advanced version of Time Warp, called Georgia Tech Time Warp (GTW), which executes in network computing environments. Because PDES systems such as GTW are essentially middleware that support their own applications, we soon realized these systems require their own middleware-specific visualization support. To this end we have extended PVaniM into a new system, called PVaniM-GTW by adding middleware-specific views. Our experiences with PVaniM-GTW indicate that these enhancements enable one to better satisfy the needs of PDES middleware than general-purpose visualization systems while also not requiring the development of application specific visualizations by the end user.

Visualizing parallel simulations that execute in network computing environments

Abstract Parallel discrete-event simulation systems (PDES) are used to simulate large-scale applications such as modeling telecommunication networks, transportation grids, and battlefield scenarios. While a large amount of PDES research has focused on employing multiprocessors and multicomputers, the use of networks of workstations interconnected through Ethernet or ATM has evolved into a popular effective platform for PDES. Nonetheless, the development of efficient PDES systems in network computing environments is not without obstacles that severely degrade simulator performance. To better understand how these factors degrade performance as well as develop new algorithms to mitigate them, we investigate the use of graphical visualization to provide insight into performance evaluation and simulator execution. We began with a general-purpose network computing visualization system, PVaniM, and used it to investigate the execution of an advanced version of Time Warp, called Georgia Tech Time Warp (GTW), which executes in network computing environments. Because PDES systems such as GTW are essentially middleware that support their own applications, we soon realized these systems require their own middleware-specific visualization support. To this end we have extended PVaniM into a new system, called PVaniM–GTW by adding middleware-specific views. Our experiences with PVaniM–GTW indicates that these enhancements enable one to better satisfy the needs of PDES middleware than general-purpose visualization systems while also not requiring the development of application-specific visualizations by the end-user.

Wireless trading in B2B markets: concepts, architecture, and experiences

With the tremendous advances in hand-held computing and communication capabilities, and the rapid proliferation of mobile devices, we are seeing a growth in mobile commerce in various consumer and business markets. In this paper, we present a novel architecture for end-to-end mobile commerce applications. We designed, implemented and deployed a system for mobile commerce connected to eMarketplaces. The system is currently undergoing trials under various configurations and in various countries. A number of end-to-end configurations for mobile commerce were experimented with, and the pros and cons were investigated. During field trials, which lasted many months, to our surprise, with WAP and HDML based clients, we were able see very good end-to-end performance for a mobile user trading at a B2B server located far away. In this paper, we also present the core issues around providing disconnected services by empowering the mobile clients with commerce functionality. Our approach is based on state-machine modeling and synchronization of objects manipulated on the client and server side.

Tools and Auxiliary Subsystems in PVM

The PVM system is a software substrate for high performance heterogeneous distributed computing on networks of workstations, mainframes, and hardware multiprocessors. PVM itself is similar to a kernel in that it provides functionality for basic operations such as process management and message passing. To augment these core facilities, a large number of tools and auxiliary subsystems have been developed, for a variety of functions such as profiling, visualization, support for other programming paradigms, input and output, etc. In this paper, we describe three such subsystems that enhance the functionality of PVM. PVaniM provides online and postmortem visualization of PVM programs, and is designed specifically for network computing environments. The PIOUS framework supports parallel and distributed I/O to facilitate applications with large data handling requirements. PVM-RPC supports the remote procedure call model to augment message passing in PVM. These tools and extensions to PVM are described, and the nature and effect of their enhancements to PVM are discussed.