Foreword

I wish I had written this book. I don ‚ t think I could have done it as well and as thoroughly as K Gopalakrishnan (Gopal), Kirtikumar (Kirti), and Richmond, but I certainly had the opportunity to some years ago. With this book available, and with its contents being so thorough, I don ‚ t think there ‚ s a need for a second book on the Oracle Wait Interface ‚ ever. At least I won ‚ t be writing one, that ‚ s for sure.

(If you somehow have managed not to hear about, read about, or work with the Wait Interface, I suggest you start with Chapter 1, which is a good introduction to the topic.)

In the book Oracle Insights: Tales of The OakTable (APress, 2004), there are some historic ‚“fun facts ‚½ about how the Oracle Wait Interface came out of obscurity and into the mainstream, mainly told by Anjo Kolk, Cary Millsap, and me, so I ‚ m not going to repeat that here.

Instead, I would like to take you on a very personal trip down Memory Lane ‚

Before I read the now-famous YAPP paper by Shari Yamaguchi and Anjo Kolk, a guy working out of Gold Support in Oracle France, Kyle Hailey, had told me in an intra-Oracle e-mail about the wait stuff and some clever tricks he was using it for, but I didn ‚ t really catch the full impact of it. (Kyle, by the way, is one of the technical reviewers of this book and is credited in one of the appendixes with his direct memory access code, and justly so.)

After reading the YAPP paper, which first championed the teaching of this method to the public, I started talking about it to everyone who cared to listen.

Whenever I had questions I would ask Anjo (we both worked at Oracle at that time), and he would usually respond within a few minutes, regardless of the time of day. In fact, I think our first contact happened a few hours after he and his family relocated from Japan to Oracle HQ, and his two sons were still suffering from jet lag.

Soon I was giving small presentations in Denmark. Then bigger presentations in Denmark. In Denmark, people will still refer to it as ‚“Mogens ‚ beloved Wait Interface, ‚½ as if I had invented it.

Then came the presentations at various user group meetings and conferences. I couldn ‚ t talk about much else in that period of my life, and I still find the whole idea and method incredibly compelling.

I had the good fortune of being used by Lex de Haan to present Oracle Technical Seminars in many wonderful countries for a couple of years, and whatever the topic was (7.3 New Features or whatever), I always made sure the audience was also introduced to the YAPP method.

I think I presented technical seminars in close to 30 different countries, and so they heard about this wonder in Argentina, Taiwan, South Africa, Italy, and anywhere in between.

One memorable occasion was presenting at a symposium arranged by Cary Millsap in his then-role as head of Oracle ‚ s System Performance Group (SPG) in Las Vegas (in 1998 or 1999, I think). Eighty people attended my presentation (the maximum allowed) and several others wanted in ‚ what a great feeling!

After the presentation, Cary strolled up to me and casually remarked, ‚“I think that was the best presentation I ever saw. ‚½ Imagine hearing that from one of your absolute heroes. I could have done anything in Las Vegas that night. But I didn ‚ t.

Fired up like that, it ‚ s hardly surprising that I kept talking about the Oracle Wait Interface whenever I had the chance.

Surprisingly, I was still one of the very few, if not the only one, to talk about the Wait Interface at the IOUG conference in Orlando in 2001. The maximum number of attendees allowed into the room was 400 and still 150 more wanted in.

That presentation went extremely well. Cary Millsap was taking care of my slides, the audience was in a very good mood, and the whole thing rocked. In fact, a nice guy in a wheelchair in the front of the room laughed so hard that at one point his glass of water fell to the floor.

During the same year, Gaja Krishna Vaidyanatha and Kirti wrote a fine performance book, Oracle Performance Tuning 101 (McGraw-Hill/Osborne, 2001), in which Gaja introduced the unforgettable phrase ‚“Compulsive Tuning Disorder. ‚½ He should also be credited with spreading the word about the Wait Interface with great enthusiasm .

I didn ‚ t participate at the recent IOUG conference in Toronto (2004), but there were a lot of presentations about the Wait Interface. The Oracle Wait Interface has truly entered the mainstream of Oracle knowledge, and I must find other things to talk about.

Cary Millsap and Jeff Holt of Hotsos have also written a very fine book, Optimizin g Oracle Performance (O ‚ Reilly, 2003), on the topic of Oracle performance methodology (including a chapter on queueing theory, which was invented in 1909 by a bored Dane, Agnar Erlang, who worked for the phone monopoly ‚ one day I ‚ ll understand it).

So with Oracle Wait Interface: A Practical Guide to Performance Diagnostics & Tuning fully (and I mean fully !) documenting the Wait Interface, its pitfalls and joys, and many tips and tricks, I think the topic is covered.

This is why I have recently taken up talking to SQL Server audiences about the virtues of a Wait Interface. And you know what? They ask me whether there ‚ s a book available on the topic.

The timing of this book is perfect. Any other time would also have been perfect if it wasn ‚ t for one important detail: With Oracle Database 10 g , Oracle (really Graham Wood, the chief architect of the Manageability area in Oracle Development and another technical reviewer of this book) is finally taking full advantage of the Wait Interface. Oracle is recording the right things, storing them correctly in a repository, and using advisory services and other utilities to use the information to its fullest.

Gopal, Kirti, and Richmond have made sure that their book contains essential details about the usage of the wait information in Oracle Database 10 g , while at the same time describing how to use it, what to look out for, and how to understand its output from Oracle 7.0.12 and up to the present. A very impressive piece of work.

The promise of all this is that optimization can be automated, but that ‚ s a bit out in the future. Until such time when we can all safely forget about performance problems, this book ought to be the preferred reference on this topic.

If it weren ‚ t for the fact that I get a copy for free, I would buy this book.

Denmark, May 2004
Mogens N ƒ rgaard
Technical Director
Miracle A/S

Acknowledgments

Never did I imagine that writing a book could be this involved. It required so much sacrifice, not only on my part, but for everyone who was directly and indirectly involved with it. I am so thankful for my wife, Jody (my crown and joy), for her help and understanding while I spent many nights and weekends clicking away on the keyboard.

I am blessed with excellent co-authors: Kirtikumar Deshpande and K Gopalakrishnan, who readily shared their knowledge and real-life experiences. Their contributions extended the depth and breadth of this book. I am also thankful for the world-class technical editors: Scott Gossett, Graham Wood, Kyle Hailey, John Kanagaraj, and Craig Shallahamer for their invaluable technical contributions and corrections. Scott, I can still vividly remember my first DBA-I class in Chicago. You were an excellent instructor and mentor.

Thanks to my outstanding superiors at Sprint Corporation: Rich Carner and all the managers in Data Management (thank you, John Mueller, for hiring me), David Hanks, David Courtney, and Michael Rapken for their support, encouragement, trust, and management style that fosters innovation. I appreciate the fine DBAs with whom I am privileged to work and share production issues and challenges. Thanks for accepting and using Oracle Wait Interface when I proposed it as a corporate standard.

Finally, thanks to the team at Osborne: Lisa McClain, Athena Honore, Jenn Tust, and Sally Engelfried for their work in preparing and making this book a reality. Although it has been hard work, I have thoroughly enjoyed it and learned much along the way.

‚ Richmond Shee

My sincere thanks go Gaja Krishna Vaidyanatha, my friend and the co-author of Oracle Performance Tuning 101 . That was the first book that promoted the use of Oracle Wait Interface methodology for Oracle Performance diagnostics. Based on the feedback we received it was apparent that there would be a book on OWI. And here it is. If it wasn ‚ t for Gaja I would never have gotten involved with writing books.

As practicing DBAs, Richmond Shee and K Gopalakrishnan have worked extensively with OWI. I am very fortunate to have them as co-authors of this book. They have contributed enormously to this book and I thank them for their invaluable contributions.

Sincere thanks go to our technical editors: Scott Gossett, Kyle Hailey, John Kanagaraj, Craig Shallahamer, and Graham Wood for their help in making sure the technical contents are correct and appropriate.

It was great working with the people at Osborne: Lisa McClain, Athena Honore, Sally Engelfried, and Jenn Tust. Thank you all very much for being so patient with us and for making sure that this project stayed on track and on time.

I would also like to thank my boss Paul Harrill for supporting and encouraging me in this effort. Without his help I am afraid that I would not have found time to work on this book.

I am thankful to everyone who contributed to this book either directly or indirectly. During the past few years I have learned a lot from many people, and my special thanks go to Steve Adams, Wolfgang Breitling, Rachel Carmichael, Dave Ensor, Daniel Fink, Tim Gorman, Anjo Kolk, Jonathan Lewis, Cary Millsap, Mogens N ƒ rgaard, Marlene Theriault, Jared Still, and many members of the oracle-l list.

Finally, I must thank my wife, Achala, and our son, Sameer. Both of them have been the true source of support and motivation. I am grateful for their support and understanding when I could not be with them while working on the book.

‚ Kirtikumar Deshpande

It all started during Oracle World 2003. When Kirti asked me whether I would be interested in co-authoring an Oracle Press book on Oracle Wait Interface, it seemed as if some mutual friend had told him of my desire . I immediately agreed but soon realized writing was not so easy after all. Without support from Kirti and Richmond, I doubt that I could have completed the chapters so quickly and so completely. I am grateful to both of them for their invaluable help.

I sincerely thank Steve Adams for getting me interested and rousing my extreme curiosity about Oracle internals, and I thank Vijay Lunawat of Oracle Racpack for directing me in the maze of RAC internals. I wish to express my heartfelt thanks to Vivek Marla, Vice President of Oracle Solution Services, India, for his encouragement and support. I am also grateful to Tarakesh Tummapudi and Muthuvel Arumugam for their valuable guidance. Many thanks to my colleague Jaswinder Singh and my friends at Oracle Support, Sudhi, Vijay, Ram(s), Sunil, and Koushal, for their timely responses to my queries ‚ mostly at odd hours. You guys are simply great!

Needless to say, any technical work needs to be reviewed thoroughly. John Kanagaraj, however, was much more than a reviewer. An Oracle guru himself, he has been an inspiration to me for the past few years. I salute him for his patience and guidance. John kept my sprits up whenever I was running low.

I am truly indebted to the esteemed clients with whom I worked. They not only posed challenges, but also were willing to implement my solutions in their systems. They played a big role in the knowledge acquisition and enhancement that I shared in this book.

Last but not the least, I wish to thank reviewers Scott Gossett, Graham Wood, Kyle Hailey, and Craig Shallahamer, whose great efforts are reflected in the rich content and timely completion of the book. Thanks, too, to Lisa McClain, Jenn Tust, Athena Honore, Sally Engelfried, and the rest of the team from Osborne for their superb work.

‚ K Gopalakrishnan