Optimizing Linux Performance

I've removed advertising from most of this site and will eventually clean up the few pages where it remains.

While not terribly expensive to maintain, this does cost me something. If I don't get enough donations to cover that expense, I will be shutting the site down in early 2020.

If you found something useful today, please consider a small donation.

I found a lot to like here, but I do need to complain about a few things. We'll do the good stuff first: there's excellent attention to detail often overlooked in performance books, the explanations are clear and illuminating, and the case studies are well done and quite interesting.

Too often a book like this will show a tool like vmstat, explain part of the output, and simply ignore the more esoteric or less interesting parts of the data. This can leave the reader wondering what certain headers mean, and send them scrambling to the man page. All too often the man page comes up short also, or gives a terse explanation that really isn't helpful. This book does better than most in that regard. I'd always like too see more, but this did fill in some gaps that man pages and other books don't bother with.

There's good attention given to explaining the relationships of data. For example, the author points out that if the number of context switches are much more than the number of timer interrupts, this must be because of sleeps or I/O waits. That's "obvious", but is easily lost in the confusion of looking at too many things at once.

The case studies are excellent, but concentrate on bugs or code optimizations rather than true system optimization. I liked that there was this kind of case study, because these often are not shown in performance books, but I wished that the author had included kernel and file system tuning studies also.

One minor annoyance cropped up throughout. When showing command line examples, the author has his prompt set to the machine name and directory, so you'll see something like this:

[[email protected] manuscript]$ sar -w -c -q 1 2
 

As he runs commands on different machines, the next example might be:

[[email protected] root]# opcontrol --vmlinux=/boot/vmlinux-2.4.22-1.2174.nptlsmp
 

Those changing prompts distracted my eyes and often momentarily confused me. I think it's best to stick to plain "#" and "$" in examples.

I would have also appreciated a little better typography. The major chapter section headings are in a larger font, but after that everything blends together. So you get something like this:

2.2.1 vmstat (Virtual Memory Statistics)

(text ..)

2.2.1.1 CPU Performance Related Options

(text ..)

2.2.1.2 Example Usage

(text ..)

2.2.2 top (v. 2.0.x)

(text ..)

2.2.2.1 CPU Performance Related Options

(text ..)

2.2.2.2 Example Usage

(text ..)

This uniformity makes it hard to go back and quickly find the beginning of a section.

Overall, I liked this and think it will find a permanent spot in my bookcase.

• Phillip G. Ezolt
• Prentice Hall
• 0131486829

Order (or just read more about) Optimizing Linux Performance  from Amazon.com

More Books

If you found something useful today, please consider a small donation.

Got something to add? Send me email.

1 comment

Have you tried Searching this site?

This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more.

Contact us