Treffer: Multi-Channel Framework for Web Application Servers: Expanding the Boundaries of Parallel Computing to Web Programming.

Web application servers are software engines that deliver applications to client computers and/or devices. They are distinguished from web servers by the extensive use of server-side dynamic content and frequent integration with database engines. In order to bring true performance (speedup and throughput) to Web application servers, we propose expanding the boundaries of parallel computing to the arena of Web Application development/programming. In this paper, we are bringing into the Web world ideas such a functional decomposition and data level parallelism from MPI; explicit concurrency directives front OpenMP, Divide-and-conquer Master-Slave and pipeline processing from parallel programming paradigms; Caching and Buffering from Parallel I/O; Fault Tolerance from PVM; Dynamic Load balancing through calibrators from Cluster and Grid Computing; and managing heterogeneity from Cluster and Grid Computing. Recently, clustering web application servers has become a reality, as demand for performance has increased, and back-end applications have become more complex. Over the past few years, recent efforts have been made on two fronts, mainly, clustering the web front-end based on web transaction dispatching such as those provided by Highly Available Load Balancing Apache Cluster, and distributing the back-end execution through the introduction of service based execution by using distributed execution mechanisms such as those provided by EJBs (Enterprise Java Beans), and Grid Services. Our proposed framework is mush more general than the above limited efforts. It allows brining many of the well tested parallel computing ideas into the Web Application World and still confirming to current Web standards. Parallel computing operates on the principle that large application can almost always be divided into smaller ones, which may be carried out concurrently. It exploits various levels of parallelisms at the task level, data level, and thread level. It has been used for many years, mainly in high performance computing, but interest in it has become greater in recent years due to physical constraints preventing frequency scaling. Parallel computing has recently become the dominant paradigm in computer architecture, mainly in the form of multicore and multithreaded processors. We argue that parallel computing will enjoy similar popularity when it extends its boundaries to the Web world. EJB testify to this but only within the Java environment. [ABSTRACT FROM AUTHOR]

Copyright of International Review on Computers & Software is the property of Praise Worthy Prize S.r.L. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)