Grinshpan L. Solving Enterprise Applications Performance Puzzles. Queuing Models to the Rescue

Издательство IEEE Press / John Wiley, 2012, -251 pp.In this book we build and analyze enterprise application queuing models that help interpret in human understandable ways happenings in systems that serve multiple requests from concurrent users traveling across a tangle wood of servers, networks, and numerous appliances. Models are powerful methodological instruments that greatly facilitate the solving of performance puzzles. A lack of adequate representation of internal processes in enterprise applications can be blamed for the failure of many performance tuning and sizing projects.
This book establishes a model - based methodological foundation for the tuning and sizing of enterprise applications in all stages of their life cycle within a corporation. Introduced modeling concepts and methodology visualize and explain processes inside an application, as well as the provenance of system bottlenecks. Models help to frame the quest for performance puzzle solutions as scientific projects that eliminate guesswork and guesstimates. The book contains models of different enterprise applications architectures and phenomena; analyses of the models that uncover connections, and correlations that are not obvious among workload, hardware architecture, and software parameters.
In the course of this work we consider enterprise applications as entities that consist of three components: business - oriented software, hosting hardware infrastructure, and operating systems.
The book’s modeling concepts are based on representation of the complex computer systems as queuing networks. The abstract nature of a queuing network helps us get through the system complexity that obstructs clear thinking; it facilitates the identification of events and objects, and the connections among them that cause performance deficiencies. The described methodology is applicable to the tuning and sizing of enterprise applications that serve different industries.Queuing Networks as Applications Models
Building and Solving Application Models
Workload Characterization and Transaction Profiling
Servers, CPUs, and Other Building Blocks of Application Scalability
Operating System Overhead
Software Bottlenecks
Performance and Capacity of Virtual Systems
Model-Based Application Sizing: Say Good-Bye to Guessing
Modeling Different Application Configurations