Customer Reviews

Optimizing Oracle Performance

5 star:		(21)
4 star:		(7)
3 star:		(0)
2 star:		(1)
1 star:		(0)

 

 

It's surprising how often people say things without knowing what they're talking about. That's a good, short description of almost everything that's been said about Oracle tuning over the last 15 years.

Surprisingly, the facts about tuning Oracle are pretty simple. Most of the results you will achieve come from doing one thing -- tuning individual SQL statements.

The author of this book is one of the few Oracle experts around, and he has created a product called HOTSOS, which is essentially an enhanced version of TKPROF. At times this book reads like the world's best infomercial, designed to promote the author's product.

Still, 99% of Oracle professionals will benefit from this book, even if you don't buy the author's product. It explains very clearly how to DIAGNOSE Oracle performance problems, and it shows an entire approach to tuning that's invaluable.

By the way, there's absolutely no need to know calculus or any other advanced math to understand this book. This rumor apparently got started by some blowhard, pretending to be an expert, but mainly interested in self promotion and telling the world, "I know calculus!" In fact, anyone with basic knowledge of SQL could understand this book and benefit from it.

In a style that will be appreciated by the beginner, intermediate and advanced DBA/Developer, Cary and Jeff introduce the reader to the world of the Oracle wait interface. This book has depth, the contents are full of meat, and it's not a rehash of every other Oracle book published in the last 10 years. In these pages is something for everyone. Cary introduces the wait interface, and how it can be used for tuning. The book presents the topic in a clear and concise fashion.

The book is full of detail such as concise information on the internal workings of Oracle trace files and how to use them to your advantage when tuning your Oracle database. They demonstrate the power of the 10046 event and it's potential.

Putting all of this together into a performance tuning method he calls Method-R, they lead us from tuning by guess work into a world of facts and using accurate measures to generate appropriate tuning responses.

This book IS required reading for any DBA who wants to be elevated to the level of a tuning deity. I've already dog eared a goodly number of pages!

This book is a great starting point for beginners (like myself) or a good reference for even experienced Oracle DBAs. It is just what you would expect from an O'Reilly book, it is very readable and presents information in a logical manner. It is refreshing to finally find a book about Oracle performance that does more than provide a collection of "tips" that you can try and then cross your fingers hoping it will improve things. I enjoyed this book thoroughly and it will definitely stay in my library.

I give it only 4 stars instead of 5 because of a couple reasons.
1. The author really pushes his hotsos profiler, which I don't mind too much. What did bother me a little bit is that in his case studies and any time showed "Method R" in action, he used the output of the profiler as a starting point. It would have been nice to assume the reader is not going to purchase the profiler and proceed from that point instead of always starting at the resource profile which the profiler outputs. This is a little nitpicky I admit.

2. I would have liked to see an expanded section on seeing Method R in action, or even downloadable sql scripts that you could run on a database and then find the bottleneck in its execution.

Overall, I definitely would recommend this book. It is well worth the money and time.

If you want a quick checklist of whats wrong with your Oracle systems - do not buy this book.

If you want validation of your existing tuning techniques - so not buy this book.

If you want to be told what to do without thinking - do not buy this book.

If on the other hand you are fed up with received wisdom that does not work, you recognize that performance tuning is a valuable skill for your business and you wish to learn an effective scientific and repeatable approach to improving the performance of your business systems, don't just buy this book but live it. Cary and Jeff take a rigorous, scientific approach to performance tuning that is especially suited to Oracle systems and by both formal proof and real world example show that the mystery of Oracle performance is controllable by you.

The only downside comes from explaining to your boss that the thousands you spent on automated performance tools were wasted.

A word about the Math. 2/3 of the book has no math at all in it. The remainder of the book was perfectly understandable to this reader, whose academic background is Economics and Philosophy not maths or computer science. Don't be put off by the idea that the book has equations, a competent High School student can follow the relevant mathematical arguments.

I purchased this book and did not open it for about 4 months. After reading the first 25 pages or so I became motivated to read more and finish it. It took me about 2 weeks to read through the entire book. The author does a really good job of detailing how to use Oracle tracing to troubleshoot performance related issues. With no shortage of Oracle performance tuning books this is definitely one of the better ones. This book must be one of the best performance related books because you can read it cover to cover. The book is packed with detail so you will reread several pages a few times, kind of like being in college and spending a couple of hours reviewing a couple of pages out of Calculus book.

If you are at all serious about becoming better at Oracle tuning this is a great book to read.

Who are the authors?

Cary Millsap and Jeff Holt were scientists at Oracle and now in their own company.

Description

The books has three parts. The first hardly talks about Oracle at all. Millsap ironically summarizes how people usually optimize, and then how to do it methodically. All Millsap is doing is reminding us to be methodical. The main lessons are: the metric that counts is user wait time, improve the slowest thing first, and calculate whether you are really saving money.
The second part speaks about Oracle. Its message is: Base yourself on the extended SQL trace data. The last chapter of the second part is a good introduction to queuing theory.
The third part shows how to put the knowledge into practice with worked examples and case studies.

Comment

The book focuses on the optimization process. It tells you how to find the right problem. It does not focus on how to fix the problem. Acceptable, because fixing is usually much easier.
The book is excellent. A PhD should be like this: readable, innovative, and useful. The book has methodic scientific attitude. This book changed my way of thinking about performance optimization.

Who should read it?

The DB admins should read it. In addition, the first part should be read by the whole team: hardware purchaser, software architect, programmer. The optimization strategies apply to optimizing anything, be it Oracle, Java, or going shopping.

Are you really reading this review because you haven't read this yet? Maybe, like me, you are just gloating. This is a once-in-a-decade read.

It went like this. Read the book. Had the epiphany. Then, as instructed read "The Goal". Epiphany II.

The book presents an optimization method, derived from "first principles". The method is exemplified in sufficient detail (albeit with simplifications) for the reader to implement.

Perhaps most importantly, it is presented in a manner that stimulates the reader on to further inquiry. With a little imagination (and this book does enourage you to use your imagination) the underlying philosophy can be extended to just about any other task you find yourself doing in the data centre.

Lucky for us that the target audience is restricted to those people interested in optimizing Oracle performance or the cat would really be out of the bag.

Another reviewer mentions "this book fills in all the gaps in analyzing Oracle performance that exist in other published works", and I disagree. This book creates gaps in most other Oracle performance analysis books. This is a tremendous book.

Some difficult material that is covered with detailed and simple explanations. If you can read and follow through with the examples you will definately come to understand even these parts.

The gaps Millsap and Holt create in other Oracle performance tuning books are 'assumptions', 'generalizations' and 'inaccuracy' because this book teaches the reader how to precisely identify user's Oracle performance problems, and exactly what to fix.

Having read this book in depth, I'll continue by saying this book also shows how to fix Oracle Performance problems. But, I almost guarantee if you are not prepared to learn the material, and follow the Method R, you may disagree this book shows 'how'.

This book is five stars because its the only reference currently available (first of class in my opionion) that provides 100% business value to Oracle performance tuning of existing systems (Oracle 7 through 10). Experience Oracle DBAs and Developers may have problems with this book because it will cause you to think about Oracle Performance in a new, refreshing way by directly improving user's response times.

Hopefully, you will read this book with an open mind and capture why Method R is so powerful to solve all sorts of Oracle performance tuning problems. Great job Millsap and Holt.

This book delivers on its promise to open up a new world of analysis and tuning to most DBAs. The concepts, calculations, and methods it provides are excellent. As an experienced tuning consultant I had used some of these methods in the past but this book brings them all together.

The book is written in a style which is enjoyable, scattered through with "war stories" and experiences which illustrate good points.

This book is not just another script book. The author not only shows you how to tune using his method but also explains why that method works. There are complex calculations included in the descriptions but it is not necessary to know higher math to use this process. No matter what method you use for tuning SQL and Oracle instances, there is material in this book which will enhance your understanding and your tuning toolkit.

There are certain "camps" in the worldwide Oracle community. For example, there is the "Oak Table Network" of "Oracle scientists" who seek thorough understandings of issues backed up by details, tests and proofs. Contrasting is the "Silver Bullet" family of field-tested generals who prefer rules of thumb and quick fixes even it means some false understandings and occasionally being wrong. Cary Millsap (of the Oak Table Network) stands as someone respected by both sides.

Cary Millsap worked at Oracle for 10 years on system performance before co-founding Hotsos in 1999 [...]. He is one of the most trusted sources on matters of Oracle system performance, and "Optimizing Oracle Performance" is considered his finest work (4.5 out of 5 stars on Amazon). The best way to learn more about him is to see for yourself. Here are some of his most popular articles:

"Diagnosing Performance Problems" from Oracle Magazine. A brief summary of what is covered in this book[...]

"Introduction", the first chapter from "Optimizing Oracle Performance."Chapter 1: [...]

"Case Study", the 12th chapter from "Optimizing Oracle Performance."Chapter 12 (Case Study): [...]

"Performance Management: Myths and Facts." One of his most popular articles.[...]

"Why a 99%+ Database Buffer Cache Hit Ratio is Not Ok." Another of his more popular articles.[...]

While everyone will have their own favourite parts of the book, I think most readers would agree that getting a good taste of the author's performance tuning philosophy is one of the highlights. "Method R", not to be confused with "System R" (ie. SQL), is not about looking at STATSPACK, cache hit ratios, or V$ tables and guessing. The author wanted to devise a system to identify and resolve the top performance concerns of an organisation with reliable, predictable results. The first few chapters put this method in writing in perhaps the best way since the introduction of "YAPP" (Anjo Kolk).

"The performance enhancement possible with a given improvement is limited by the fraction of the execution time that the improved feature is used." - Amdahl's Law

After several years of research, the author discovered that Extended SQL Trace Data was at the centre of "Method R". Some of the articles should give you a good taste of what Extended SQL Trace data is, if you didn't know already. By the time you finish reading this book you will know exactly how to collect and interpret all the little "ela=17101 p1=10 p2=2213 p3=1 ..." within into something meaningful. For some, that justifies the price tag right there.

So in essence I would have re-named this book "Method R: Optimizing Oracle Performance Using Extended SQL Trace Data," because that is basically what this book is about. There are some reasonably "stand-alone" chapters on other topics, for instance on the Oracle Fixed View tables (Chapter 8) and on Queueing Theory (Chapter 9), but that is not the primary focus of the book.

Those that are expecting a more broad treatment of the subject of performance tuning may be justifiably disappointed that it basically covers only this narrow aspect. However, it is covered very well, and it isn't really covered anywhere else. The author makes no apologies for this, claiming that extended SQL trace data is the only resource you will ever need for diagnosing and solving performance problems.

"You cannot extrapolate detail from an aggregate." - Cary Millsap's preference of SQL extended trace data over fixed views (system-wide average performance characteristics)

Indeed, some people might contend that the author spends a little too much time stating his beliefs, defending them, and patting himself on the back. But I think it adds a certain flavour to the book, and I respect an author who backs up his statements.

"Proving that V$ data are superior to extended SQL trace data because of the 'missing time' issue is analagous to proving that its safer to be in a room with a hungry bear if you'll just close your eyes." - Cary Millsap

The book can be a tough read in the sense that the author goes very deep into the material, and generally each subject is treated thoroughly. Chapter 9 on Queueing Theory can be a particularly overwhelming chapter. But the material is served in bite-size pieces, and broken up with tips, tricks, stories, diagrams and code (sometimes 3+ pages worth at a time, embedded directly in the middle of a chapter). There are even worthwhile exercises at the end of each chapter.

In the end, I enjoyed this book and I'm glad I got it. I don't consider it a "must have" for your Oracle collection, but I definitely feel it is quite worthwhile. I recommend it especially to those who read his articles and were very comfortable with his writing style and philosophy, and also to those that need a book on extended SQL trace data (because this is basically the only one). But even those in the "Silver Bullet" camp will be glad to add another tool to their belt.

Thumbs up.