Performance Assurance for IT systems

Citation

Moscato, D.R. (2005), "Performance Assurance for IT systems", Benchmarking: An International Journal, Vol. 12 No. 3, pp. 283-284. https://doi.org/10.1108/14635770510600393

Performance assurance for IT systems provides for a welcome addition to the literature on building quality IT systems. Its unique contribution lies in its concentration on the pre‐production stage of the project life cycle. It is a useful complement to the existing body of work that emphasizes performance tuning and other post production issues. A distinct advantage of this book is that the author, Brian King, is not hesitant to share his varied experiences with the reader in the form of tricks and checklists.

The organization of the book adds to its usefulness as both a quick reference to the role of quality assurance during each stage of the project life cycle as well as a more complete reference to the underlying technologies that support the quality assurance effort in organizations. The book is divided into two parts. The first part (184 pages) presents the project life cycle framework and how quality assurance can impact each stage. The second part of the book (150 pages) provides the reader with a collection of “tasters” (author's own word) on the technology that supports the underlying concepts discussed in the first part of the book. The reader who has a sufficient technical background can skip over part two and concentrate on the main contribution of the book. Although I am very comfortable with the technical foundations presented by the author, I found his coverage of the topics to be very informative. Each “taster” followed the same five‐part format. There are sections on basic terminology, a technology outline, performance issues, general observations (often very insightful) and further reading. Since this part of the book consumes almost half of the text, the reader is able to pick and choose only those primers that are required to add to his knowledge base. The author points out that these “tasters” are not meant to be exhaustive but merely to capture the essential issues on the topic. Each technology, in his opinion, can be described as being composed of no more than six key drivers. The interested reader can select one or more of the annotated references to obtain a more complete understanding of the underlying technologies. It is important to point out that each module is extremely concise but very well‐written. King has the ability to hone in on the essential drivers of each technology factor and provide its link to the performance of the resulting IT system.

The major contribution of this book rests with part one. The author is very careful to articulate the audience and perspective for the material. His basic tenet is that performance assurance (an approach for tackling performance) should be employed proactively throughout the development life cycle. To King, the ideal foundation for the sizing and performance analyst is a solid understanding of hardware and software technologies. However, he maintains that there must also be a solid competence in the “soft side” of the process. He devotes coverage to all perspectives of the development process including the proposal stage, dealing with vendors, evaluating the financial viability of the vendor, benchmarking systems and many other topics not usually found in a text of quality assurance.

Even though the book features many of the author's rather opinionated viewpoints on virtually all of the topics, it is important to say that he refrains from direct comparisons between the vendors and any specific products. This approach is not easy to do, but the reader will appreciate his candor as it pertains to some of the practices used by vendors of both hardware and software products. When appropriate, King does not hesitate to admit that there will be times that there is a paucity of good information available to perform the kind of analysis that is required for an evaluation of an IT system. The key point is “good information”. Vendors have the ability to swamp the client with data under the guise of relevant material. His advice, if adhered to, has the potential to avoid many a catastrophic outcome. My own experience in this field parallels the author's and many of his conclusions I am able to corroborate first hand.

Overall, this book is both well‐written and edited. It contains many useful illustrations that add to the readability of the text. One of its greatest strengths is its organizing theme – “the emphasis should be on preventing problems, rather than waiting for them to surface”. By using a modified Windfall model of the development process, King is able to superimpose quality assurance principles on every stage of the process. He argues that one “should design for performance…rather than to try to retrofit performance into the system afterwards”. To him, cost is the ultimate driver of any IT system. In essence, King wrote this book because he believes that too little attention is paid to the performance during the preliminary stages of the development process. All too often, it is left to the tuning or post‐production phases. If you are wondering what the requisite skill set is that is necessary to engage in the kind of effort espoused by King, it is a “hybrid of technical architect and performance analyst.” Unfortunately, King would be the first to admit that this type of valuable person is a rare bird.

If you are actively involved in any phase of the SDLC or RAD processes, then performance assurance for IT systems should be a must read and a welcome addition to your personal or corporate library. Since it is a professional book that is written for the practitioner, it is not suitable as a principal textbook in a university environment unless the professor is willing to supplement it with exercises and other support materials.