Book Review - Highly Available Storage for Windows Servers
A few months ago I started a review of another storage book by observing that nearly all the major publishers now have a select few storage titles on the shelves. Most of these are general storage networking books designed to appeal to as wide an audience as possible. As the storage industry develops, though, and as tech's become more storage savvy, specialized titles are also starting to appear in numbers. One such book, which has actually been around for almost a year, is Highly Available Storage for Windows Servers, a 450 page hardback title from Wiley Computer Publishing, which takes an in-depth look at managing storage on the worlds most popular server platform.
Wiley - generally regarded as one of the more prestigious technical publishing houses has teamed up with software manufacturer VERITAS to produce what they call the VERITAS Series of which 'Highly Available Storage' is part. The general intent of the book in the series appears to be to tackle topics from a practical perspective, and discuss issues as they affect the day-to-day use of storage in working environments.
The Sum of Two Parts
The book is broken down into two distinct parts. The first part - Disk Storage Architecture - deals, in a general manner, with disk storage concepts such as filesystems, volumes, RAID and so on. The explanations in each instance are excellent but experienced storage administrators might find themselves skipping over material in the belief, which may be justified, that they don't need to read yet another explanation of these most basic storage principles. While this is a reasonable assumption, the quality of coverage in this section is better, on a technical level, than in many other books and is certainly deserving of attention from all but the most experienced of Windows administrators. Another thing that should be mentioned is that Part One of the book only accounts for some 90 odd pages of the book - the other 310 pages being dedicated to Part Two - Volume Management for Windows Servers.
A good way to describe Part Two is a bit like a guide to Disk Administrator - on steroids. Topics like management of volumes in clusters and data replication covered - always from a perspective of relevancy. The author explains not only why you might do something but also how (in detail) and when. Part Two includes five chapters on volume management, a brief look at multipath data access, a discussion of managing hardware disk arrays, coverage of managing volumes in clusters and a discussion of Data Replication.
Watch For the Pitch
In light of the fact that the book is part of the VERITAS Series, it should come as no surprise to find that there are mentions of certain VERITAS products such as VERITAS Volume Manager and VERITAS Cluster Server, but the utilities supplied by Microsoft are also covered in as much, if not more detail. Whether the discussion of VERITAS products bothers you or not will depend on your point of view. I didn't find the discussions too sales oriented, nor did I get lost about whether the discussion was about the VERITAS product or the Microsoft tools. Still, you could say that if you are paying for the book you don't need the pitch, no matter how 'unsalesy' it is made. From a different angle, if you are using one of the VERITAS products the book takes on an even greater meaning - I would wager that the information provided blows away what's available in documentation and manuals.
Some Sound Recommendations.....Twice
A chapter titled 'Windows Online Storage Recommendations' rounds off the book nicely. What you get from it is 37 'Rules of Thumb' which the author encourages you to observe when managing storage on Windows servers. Some of the rules are brief, others not, but what they represent is 37 lessons learned, apparently in the field, by some of the most experienced engineers in the storage industry. A great number of the 'recommendations' are the kind of thing you don't think about until it's too late. The advice regarding the locating of data and the recovery mechanism on separate disks so that a single disk failure does not render both the data and the recovery mechanism unavailable is a good example of this. An odd footnote to the discussion of this chapter is that the recommendations are listed again, this time without any explanation, in Appendix Two. If you just said "eh?", that was my reaction too. Seeing as Appendix Two only starts four pages after Chapter 14, I am a little confused as to the purpose of this duplication - if there is one.