Fault-Tolerant Network-Server Architecture For Time-Critical Web Applications

Keywords

fail-over; NLB; Redis; servers; STP; time-critical; VRRP; websocket

Abstract

Parallel to recent evolution in web content, technologies such as WebSockets and WebRTC allow development of real-time web applications - something that has not been possible in RESTful HTTP platforms. Future IoT applications will also require real-time data retrieval as information travels through and is processed by web servers. For these reasons, we envision a demand for real-time web applications which require a high cost to service providers in case of service interruption. A time-critical real-time application is assumed to have the capability to handle failures in infrastructure, not with a service interruption, but with minor performance degradation. In this study, a fully-redundant system architecture for web server deployment, including the network and server, with fast failover capabilities is presented. Unlike interruptions that require applications to be re-initialized, this proposed system provides fail-over speeds for every component, such that, even the fast real-time applications can continue to operate. Experiments with application scenarios prove superior failover speeds under several test cases and reveal limitations in the latest technologies in supporting the aforementioned applications.

Publication Date

3-29-2018

Publication Title

Proceedings - 2017 IEEE 15th International Conference on Dependable, Autonomic and Secure Computing, 2017 IEEE 15th International Conference on Pervasive Intelligence and Computing, 2017 IEEE 3rd International Conference on Big Data Intelligence and Computing and 2017 IEEE Cyber Science and Technology Congress, DASC-PICom-DataCom-CyberSciTec 2017

Volume

2018-January

Number of Pages

377-384

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

DOI Link

https://doi.org/10.1109/DASC-PICom-DataCom-CyberSciTec.2017.79

Socpus ID

85048087764 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85048087764

This document is currently not available here.

Share

COinS