Solving Enterprise Applications Performance Puzzles (eBook)

Leonid Grinshpan Ph.D. (Stamford, CT) has been working for Oracle Corporation for more than 14 years as Consulting Technical Director. His customers include Dell, Citibank, Verizon, Clorox, Bank of America, AT&T, Best Buy, Aetna and many other major enterprises. He obtained Ph.D. in queuing models of computer systems from Russian Academy of Science.

Acknowledgements.

Preface.

Why I wrote this book?

Subject.

Audience.

Organization.

Chapter 1. Queuing networks as application's models.

1.1 Enterprise applications - what do they have in common?

1.2 Key performance indicator - transaction time.

1.3 What is application sizing and tuning?

1.4 Queuing models of enterprise application.

1.5 Transaction response time and transaction profile.

1.6 Network of highways as analogy of queuing model.

Chapter 2. Building and solving application models.

2.1 Building models.

2.2 Essentials of queuing networks theory.

2.3 Solving models.

2.4 Interpretation of modeling results.

Chapter 3. Workload characterization and transaction profiling.

3.1 What is application workload?

3.2 Workload characterization.

3.3 Business process analysis.

3.4 Mining transactional data from production applications.

Chapter 4. Servers, CPUs, and other building blocks of application scalability.

4.1 Application scalability.

4.2 Bottleneck identification.

Chapter 5. Operating system overhead.

5.1 Components of operating system.

5.2 Operating system overhead.

Chapter 6. Software bottlenecks.

6.1 What is software bottleneck.

6.2 Memory bottleneck.

6.3 Thread optimization.

6.4 Other causes of software bottlenecks.

Chapter 7. Performance and capacity of virtual systems.

7.1 What is virtualization?

7.2 Hardware virtualization.

7.3 Methodology of virtual machines sizing.

Chapter 8. Model-based application sizing - say good-bye to guessing.

8.1 Why model-based sizing?

8.2 Model's input data.

8.3 Mapping system into model.

8.4 Model deliverable and what-if scenarios.

Chapter 9. Modeling different application configurations.

Glossary.

References.

Index.

Solving Enterprise Applications Performance Puzzles : QueuingModels to the Rescue by Leonid Grinshpan is a prettyinteresting book about application of queuing models to solvingenterprise performance and I believe the book fills a few gaps inpractical application of queuing theory. Another good name for thisbook could be "Building queuing models by example".

I spent a lot of time trying to use queuing models to solvepractical performance issues and would testify that it is prettychallenging. There are a few areas where it was developed a littlefurther (for example, around capacity planning of existingsystems), but if you trying to do something else - youwon't find much help. You have a lot of books about systemsperformance, you have a lot of books about queuing theory withsimple examples, but not much in between to solve practical tasks.And here Leonid's book may help, especially if you are new inthis area.

Chapter 1, Queuing Networks as Applications Models, is anintroduction into the topic. It discusses how queuing theory may beused to model enterprise applications. A lot of analogues are usedto introduce the subject.

Chapter 2, Building and Solving Application Models, is anoverview of the whole process, including short discussions aboutessentials of queuing theory and using of tools to solvemodels.

Chapter 3, Workload Characterization and Transaction Profiling,discusses what input data for models are and how to gatherthem.

Chapter 4, Servers, CPUs, and Other Building Blocks ofApplication Scalability, discusses scalability, bottlenecks, how toidentify bottlenecks and ways to fix them (mostly on CPU and I/Oexamples).

Chapter 5, Operating System Overheads, discusses main componentsof operating systems, where overheads come from, how to measurethem, and their impact on transaction time.

Chapter 6, Software Bottlenecks, is devoted to softwarebottlenecks, which are rarely discussed in application to queuingmodels - while in practice software bottlenecks happen allthe time. Memory bottlenecks and thread optimizations and theirmodeling are discussed in details. Multiple other softwarebottlenecks are also reviewed.

Chapter 7, Performance and Capacity of Virtual Systems, is anoverview of performance issues related to virtualization , theirexplanation with queuing theory, and a methodology of virtualmachine sizing.

Chapter 8, Model-Based Application Sizing: Say Good-Bye toguessing, explains why to use model-based sizing and discusses itstep-by-step from gathering input data to model deliverables andwhat-if scenarios.

Chapter 9, Modeling Different Application Configurations,discusses several specials cases including geographicaldistribution of users, cross-platform modeling, remote terminalservices, load balancing, and parallelization of transactions.

The book covers a lot of topics. However, to avoiddisappointments, I'd like to point out what this book isnot:

- It is not a textbook about queuing theory. The section 2.2Essentials of Queuing Networks Theory has 5 pages in it.

- It is not a book about tools to solve queuing models.Available tools are listed and there are references, but they arejust mentioned as a way to solve models (with one tool used as anillustration of the process). You don't need to know any toolto read the book (but you will need one when you try to solve yourown models).

- It is not a comprehensive book about enterprise applicationperformance. There is plenty of important information and practicalrecommendations about enterprise application performance in thebook, but it is shared as needed to build models and analyze theirresults.

So the book is exactly what the title says: a practical bookabout building queuing models to investigate enterpriseapplications performance issues.

- Alexander Podelko, Oracle