This paper discusses the design and implementation of a high-availability clustering solution.This solution provides data integrity guarantees and a simple and effective way for making services highly available.The implementation of the system design in a Linux environment on commodity hardware is documented.A proof-of-concept to provide a highly available NFS server is explained with caveats and their resolutions.
Citation:
Richard Rabbat, Tom McNeal, Tim Burke, "A High-Availability Clustering Architecture with Data Integrity Guarantees," cluster, pp.178, Third IEEE International Conference on Cluster Computing (CLUSTER'01), 2001