Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Power and performance : software analysis and optimization
Kukunas J., Morgan Kaufmann Publishers Inc., San Francisco, CA, 2015. 300 pp. Type: Book (978-0-128007-26-6)
Date Reviewed: Mar 21 2016

Brief explanations of why performance analysis and optimization are important, followed by a deep dive into the hows and wheres, are provided in this book. Even though it uses Intel architecture and the Linux operating system as use cases, it covers the intended topics with enough clarity and depth to serve both as a potential textbook and as a reference for practitioners. It is organized into three parts, each with multiple chapters.

Part 1, “Background Knowledge,” starts by covering Intel architecture history with the intent of explaining the status quo--that is, how it evolved to include new functionality, more power, more performance, and yet remain backwards-compatible. The last two chapters in Part 1 are called “Performance Workflow” and “Designing Experiments.” They are a good introduction for anyone interested in learning what it involves to do good performance work, from problem statement to testing, choosing adequate metrics, controllable versus uncontrollable parameters, and so on. An existing testing suite is used as a practical example.

Part 2 is called “Monitors” and addresses existing performance tools. Its first chapter explains performance counters and how they may be used to detect performance issues. The next five chapters in Part 2 are each devoted to a specific tool or group of tools such as VTune, Perf, Ftrace, or Graphics Profiling.

Finally, Part 3, called “Optimization Techniques,” contains five chapters. The first and longest chapter covers what the author calls “a set of tools for building, inspecting, and modifying software,” including compiler flags, some key functions, programming techniques such as using appropriate types and qualifiers, loop unrolling, and so on. The last four chapters cover branching, cache usage, parallelism, and some Intel special instructions.

This one of the best technical books I have read in a while. Having said that, there is some room for improvement on the organization front: some important terms, for example model-specific registers (MSR), are missing from the index, and a glossary would be nice.

More reviews about this item: Amazon

Reviewer:  Veronica Lagrange Review #: CR144243 (1606-0364)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Performance of Systems (C.4 )
 
 
Compilers (D.3.4 ... )
 
 
Program Analysis (F.3.2 ... )
 
 
Processors (D.3.4 )
 
Would you recommend this review?
yes
no
Other reviews under "Performance of Systems": Date
A computer and communications network performance analysis primer
Stuck B., Arthurs E., Prentice-Hall, Inc., Upper Saddle River, NJ, 1985. Type: Book (9789780131639812)
Jun 1 1985
A mean value performance model for locking in databases
Tay Y., Suri R. (ed), Goodman N. Journal of the ACM 32(3): 618-651, 1985. Type: Article
Mar 1 1986
The relationship between benchmark tests and microcomputer price
Sircar S., Dave D. Communications of the ACM 29(3): 212-217, 1986. Type: Article
Nov 1 1986
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy