Modern web-based application infrastructures are based on clustered, multi-tiered architectures, where request distribution occurs in two sequential stages: over a cluster of web servers, and over a cluster of application servers. Much work has focused on strategies for distributing requests across a web server cluster in order to improve overall throughput across the cluster. The strategies ap...

We present the Scalable, Application-Space, HighlyAvailable (SASHA) architecture for network-clustered web servers that demonstrates high performance and fault tolerance using application-space software and CommercialOff-The-Shelf (COTS) hardware and operating systems. Our SASHA architecture consists of an application-space dispatcher, which performs OSI layer 4 switching using layer 2 or layer...

The need for web services is constantly increasing, while the requirements for web services are getting more and more complicated. Thus, web server systems with ever increasing performance are constantly in need. To efficiently satiate this need, web cluster systems, which have been a topic of much research, have been suggested due to its scalability. In this paper, we design and implement a sc...

In this paper, a methodology to exploit legacy sequential libraries on a multicomputer is shown. The process of integrating such libraries is exposed as a slave pool design pattern. The proposed solution is implemented using the concurrent functional programming language Erlang, and it has been applied to extend a C++ sequential XSLT library to perform clustered XSLT processing on a Beowulf, wh...

Clustered servers for Internet service is a popular solution to cope with the explosive increase in client requests. The high probability of service failure in cluster servers make the cluster management system necessary to provide high availability and convenient administrator control. In this paper, we present the design and implementation of a load cluster management system (LCMS) based on S...