Ef .ciently distributing component-based applications across wide-area environments

Abstract

Distributed deployment of network applications in widearea environments has proven effective for improving enduser experience. Another trend is the use of component frameworks for building network services. Their component-based nature makes such applications natural candidates for distributed deployment, but it is unclear if the design patterns underlying component frameworks also enable efficient service distribution.In this paper, we investigate the application design rules and accompanying system-level support essential to a beneficial and efficient service distribution process. Our study targets the widely used Java 2 Enterprise Edition (J2EE) component platform and Java Pet Store, a sample component-based e-commerce application.Our results present strong experimental evidence that component-based applications can be efficiently distributed in wide-area environments using a small set of generallyapplicable design rules for orchestrating interactions and managing component state. We additionally discuss enforcement of these rules, and their automated implementation by container frameworks.

RDRP: Reward-Driven Request Prioritization for e-Commerce web sites

“…Since then several other studies have explored session characterization of web workloads (Menascé et al, 1999;Shi et al, 2002;Akula and Menascé, 2007). The work by Llambiri et al (2003) acknowledged that service usage patterns affect the performance of Internet services, and that understanding the nature of user workloads is crucial for properly designing and provisioning web servers. Our study follows this trend, by using aggregated information about client usage of an e-Commerce service to boost the service profit or other application-specific reward that clients bring to the service provider.…”

Exploiting Service Usage Information for Optimizing Server Resource Management

“…Second, both local and remote clients experience improved performance due to aggressive caching of stateful components. Further details about the design patterns and optimizations, their implementation, and the experiment results can be found in Llambiri et al [2003].…”

Infrastructure for Automatic Dynamic Deployment of J2EE Applications in Distributed Environments

“…The variety of deployed components resulting from the usage of our infrastructure represents an application-level overlay network of J2EE components analogous to that of [2], [7], and [14], where several instances of the same component may coexist together. We believe that J2EE limitations on component lifecycle, concurrency and state may allow for efficient models of consistency between multiple versions of the same stateful J2EE component.…”

“…Vertical partitioning (e.g., running web tier and business tier in separate VMs) has been used for fault isolation and load balancing but it is sometimes impractical due to significant run-time overheads (even if one would keep the tiers on a fast local-area network) related to heavy use of remote invocations. Recent work [14] in the context of J2EE componentbased applications has shown viability of vertical partitioning in wide-area networks without incurring the aforementioned overheads. The key conclusions from that study can be summarized as follows:…”