Clusters for High Availability: A Primer of HP Solutions (2nd Edition)

With an increasing share of the world's economy dependent on computer systems, the importance of making them work without interruption is growing. Clusters for High Availability: A Primer of HP Solutions introduces readers to computer system faults and to techniques for designing systems for maximum reliability and fault tolerance. The book emphasizes Hewlett-Packard's line of High Availability (HA) products, but also explains HA problems and solutions so as to provide value to all readers, regardless of whether they're running HP gear. The explanations begin with pretty basic concepts, so this book has value for engineers designing HA strategies from the ground up.

The best part of the book is the discussion on classes of failure and the hardware and software strategies available for dealing with each. The author points out the strengths and weaknesses of each approach, staying completely clear of explicit instructions, which allows the reader to get them from documentation after he or she has chosen a HA strategy and associated products. The book's conceptual diagrams are fun too: Several diagrams indicate the failure of a data center by superimposing giant flames over a diagram of the center's components! --David Wall

Topics covered: High availability for data centers and other computer systems via component redundancy, careful software design, and especially server clustering. Event Monitoring Services (EMS), ServiceGuard, and High Availability Observatory (HAO) are among the Hewlett-Packard solutions covered, though the author ranges beyond specific products to cover general design practices.