23 Second Kernel Compiles

b-side.org writes "As a fine testament to how quickly linux is absorbing technology formerly available only to the computing elite, an LKML member posted a 23 second kernel compile time to the list this morning as a result of building a 16-way NUMA cluster. The NUMA technology comes gifted from IBM and SGI. Just one year ago, a Sequent NUMA-Q would have cost you about USD $100,000. These days, you can probably build a 16-way Xeon (4X 4-way SMP) system off of ebay for two grand, and the NUMA comes free of charge!"

Comments

Who would have guessed...

[CmdrTaco (editor)]

that technology would progress, and previously unaffordable hardware would become cheaper and more versatile? This is not so much Linux absorbing elite technology as it is the evolution of technology in general.

For example, just a few years ago my main computer was a Intel 486 running at 100 Mhz. Just think how long it would take to compile the kernel on that system. From that I went to a Pentium 90, then a Pentium 233, a Celeron 533, a Pentium III 533, a Pentium III 800, and finally a Pentium III 933...

A good portion of my life is lived, if you'll allow me the word, on-line. It might even be said that I live a double life, part of it with family, friends, and co-workers in the immediate, material world, and another part with a different (although sometimes overlapping) circle of family, friends, and co-workers on the net. Not only am I connected to other people, but via the Internet I am also connected to a collection of tools and resources that help me in my day-to-day life. For example, I use hypertext as a thinking tool for producing and developing my ideas; I use the World Wide Web as a canvas for mapping and presenting myself and my work; and I use email not only to keep in touch, but also to manage my time and organize my projects.

I will not argue: our culture is indeed defined by our formulations of foundational concepts such as religion, family, and truth. But, with time, meaning changes. The defining aspects of a culture are not mystically immutable. On the contrary, they experience drastic changes, causing jihads and holocausts, but also rebirths and renaissances. At this turn of the century it is clear that computing and related technology is pivotal to our culture's progression, for better or for worse. And my response to the "neo-Luddites" is to ask them to pay attention to the "or." There is no evidence or sound argument that our redefinition of basic cultural terms spells out doom. Changing, evolving society-wide conceptions are a routine part of any culture's movement through time, no matter how "unnatural" the cause of the evolution might appear to be. Just as it is too soon to be heedless and inattentive to the consequences of computing and the Internet, it is also too soon to be so afraid.

In the end, the arguments of the "neo-Luddites" do not withstand deeper inquiry. Neil Postman, Clifford Stoll, and Stephen Talbott repeatedly attribute to the Internet qualities which actually describe our use of, or attitude toward, the Internet, not the Internet itself. It is clear that the technology is neither intrinsically good nor evil: it is neutral. We can have meaningful experiences with it, or waste our time. We can be lost in its flashy colors and informational floods, or we can take control. We can lose our identities, or hold on to them. There is no goodness or badness intrinsic to the Internet any more than there is goodness or badness intrinsic to a block of marble. The marble may tend toward cleaving one way rather than another, as a medium is inclined to, but its nature -- its feeling, its attitude, its content -- is left to us, not to it. With this awareness we can shape the on-line future to our best ideals. "It's all a question of what we do with it," Talbott wisely observes of the Internet. And despite the fact we differ on many points, on this point we agree. He writes:

;) [lameness filter]

[Scoria]

absorbing technology formerly available only to the computing elite but we are l33t!1111111 d0nt j0 c4ll m3 c0mpu3t3r3 l3tma4m3@!!11111 1 l33t!11111 Absorbing technology? Extend and embrace? :p

Long compile tricks

[LinuxHam]

While I may not recall my first kernel compile (late 96 or early 97), I wanted to share a small kernel compile trick. When compiling on my P75's, I usually

nice -n -20 screen
make dep clean bzImage modules modules_install ; echo "Kernel done" | Mail myname@skytel.com

I started doing that in '98 when compiling on a 486 I commandeered in the server room. Saves you from doing "are we there yet?" with your box. Also lets me know when I need to head home from running my errands or eat a little faster at the dinner table. :)