It is a characteristic of most computer systems that they do not degrade gradually; performance is acceptable day after day, until quite suddenly it all falls apart. When this happens, the administrator needs to be prepared to help the organization get through the crisis. Computer applications are growing ever more intelligent and easy to use. One of the by-products of making applications easier to use is that they usually also require more resources to run. Wherever productivity is a central factor in the decisions made, performance considerations loom large and continue to play an important role in system management. This book shows why it is important to understand the performance characteristics of the hardware and of the workload, and how they match up against each other. It takes the reader through problem-solving techniques like measurement methodology, workload characterization, bench-marking, decomposition techniques and analytic queuing models. Among the subjects the book covers are: processor performance; application profiling and hardware considerations; multi-processing; memory and paging; file cache; disk performance; IIS; and networking.
Odysseas Pentakalos has been an independent consultant for 10 years in performance modeling and tuning of computer systems and in object-oriented design and development. His clients have included major government agencies and corporations such as NASA, the Army Research Lab, Sun Microsystems, and Concert Communications. Odysseas received a Ph.D. in computer science from the University of Maryland. He has published papers on performance topics in conferences, journals, and commercial publications. Mark Friedman began his career as a programmer for the DuPont Corporation in 1977 and has been in the computer field ever since. He has a master's degree in computer science from Temple University and is the founder and president of Demand Technology Software. He has written numerous technical articles, conducts training seminars in Windows performance, and publishes a monthly newsletter. Currently, he is working on the design and development of professional software tools for Windows performance management.
Preface; 1. Perspectives on Performance Management Windows 2000 Evolution Tools of the Trade Performance and Productivity Performance Management Problems of Scale Performance Tools; 2. Measurement Methodology Performance Monitoring API Performance Data Logging Performance Monitoring Overhead A Performance Monitoring Starter Set; 3. Processor Performance Windows 2000 Design Goals The Thread Execution Scheduler Thread Scheduling Tuning; 4. Optimizing Application Performance Background The Application Tuning Case Study Intel Processor Hardware Performance; 5. Multiprocessing Multiprocessing Basics Cache Coherence Pentium Pro Hardware Counters Optimization and SMP Configuration Tuning Configuring Server Applications for Multiprocessing Partitioning Multiprocessors; 6. Memory Management and Paging Virtual Memory Page Replacement Memory Capacity Planning; 7. File Cache Performance and Tuning File Cache Sizing Cache Performance Counters Universal Caching How the Windows 2000 File Cache Works Cache Tuning; 8. Disk Subsystem Performance The I/O Subsystem Disk Architecture I/O Buses Disk Interfaces System Monitor Counters Workload Studies ;9. Filesystem Performance Storage Management Filesystems Defragmentation System Monitor Counters Comparing Filesystem Performance Selecting a Filesystem; 10. Disk Array Performance Disk Striping Enter RAID RAID Disk Organizations RAID and Windows 2000 Benchmark Testing Selecting a RAID Configuration; 11. Introduction to Networking Technology Networking Basics Bandwidth and Latency Media Access Layer Internet Protocol Layer Host-to-Host Connections; 12. Internet Information Server Performance Web Server Architecture Sources of Information Web Server Benchmarks Performance Management Load Balancing and Server Clustering; Bibliography; Index