Power and Performance

Informationen zum Autor Jim Kukunas began programming at a young age, teaching himself C and x86 assembly. He is an alumnus of Allegheny College with a degree in Computer Science. Today, he is a software engineer in Intel's Open Source Technology Center. As a performance optimization engineer on the core Linux kernel team, much of his work focuses on kernel space and user space performance optimizations. His efforts have enhanced many projects including the Linux kernel, Zlib, the Englightenment Foundation Libraries, Meego, Android, and many others. Klappentext Power and Performance in Enterprise Systems is a guide to solving performance problems in modern systems. Beginning with the necessary architectural background as a foundation! the book demonstrates the proper usage of performance analysis tools in order to pinpoint the cause of performance problems! also including best practices for handling the common performance issues that those tools identify. . The text presents ideas to improve architectures running mobile! desktop! or enterprise platforms and explains the importance of parallelism and profiling to determine the source of performance issues. It finishes with case studies that highlight how all of these methodologies are combined to produce measurable results.

Autorentext
Jim Kukunas began programming at a young age, teaching himself C and x86 assembly. He is an alumnus of Allegheny College with a degree in Computer Science. Today, he is a software engineer in Intel's Open Source Technology Center. As a performance optimization engineer on the core Linux kernel team, much of his work focuses on kernel space and user space performance optimizations. His efforts have enhanced many projects including the Linux kernel, Zlib, the Englightenment Foundation Libraries, Meego, Android, and many others.

Klappentext

Power and Performance in Enterprise Systems is a guide to solving performance problems in modern systems. Beginning with the necessary architectural background as a foundation, the book demonstrates the proper usage of performance analysis tools in order to pinpoint the cause of performance problems, also including best practices for handling the common performance issues that those tools identify. . The text presents ideas to improve architectures running mobile, desktop, or enterprise platforms and explains the importance of parallelism and profiling to determine the source of performance issues. It finishes with case studies that highlight how all of these methodologies are combined to produce measurable results.

Inhalt

Part I: Background Knowledge1. Early Intel Architecture2. Intel Pentium processors3. Intel Core processors4. Performance Workflow5. Designing Experiments

Part II: Monitors6. Introduction to Profiling7. Intel VTune Amplifier XE8. Perf9. Ftrace10. GPU Profiling Tools11. Other Helpful Tools

Part III: Optimization Techniques12. Toolchain Prime13. Branching14. Optimizing Cache Usage15. Exploiting Parallelism16. Special Instructions