Solving Enterprise Applications Performance Puzzles: Queuing Models to the Rescue

From Performance Chaos to Queuing Clarity

Enterprise applications often deliver unpredictable response times, unexplained bottlenecks, and capacity surprises. Leonid Grinshpan cuts through that complexity with a direct, model-driven approach built on queuing theory. The book shows how to construct practical models that reveal why systems behave the way they do—and what will happen when hardware, workload, or architecture changes.

What the Book Covers

The material is organized around the skills performance engineers need most:

• Building and calibrating queuing models of enterprise applications
• Characterizing workloads and mining transaction data
• Identifying CPU, I/O, memory, and thread-related bottlenecks
• Analyzing virtualized environments and sizing virtual machines
• Modeling remote offices, load balancing, and parallel processing

Real-World Problems, Not Just Theory

Grinshpan draws on years of hands-on Oracle consulting to connect mathematical concepts with everyday performance troubleshooting. Each chapter builds a toolkit for diagnosing slowdowns, predicting scalability limits, and testing “what-if” scenarios before committing to expensive changes. The writing is practical enough for engineers who need results, yet grounded enough for those who want to understand the underlying principles.

A Sharper Approach to Capacity and Tuning

The book addresses common weaknesses in ad hoc performance work: guesswork about user concurrency, incomplete workload profiles, and unclear metrics. By translating enterprise systems into queuing networks, readers gain a repeatable way to separate real limitations from red herrings—and to communicate findings with confidence.

Who Will Find It Useful

Performance engineers, capacity planners, system architects, and database administrators working with enterprise applications will find immediate value. The material is also suitable for software developers who want to anticipate performance issues during design, and for IT managers who need to understand the trade-offs behind serious performance investments.

Take the Guesswork Out of Performance Work

If you have ever struggled to explain why an application slowed down or how many users a new server can really support, this book offers a rigorous yet accessible path forward.