What's most charming about collective computing is that it
relies entirely on resources that would otherwise go to
waste. The computing is done with spare CPU cycles on idling machines.
It is one of the wonders of our age that we squander vast
quantities of computational labor. Forty years ago, when
electronic computers were rare and expensive, CPU time was
scheduled and billed by the millisecond. Now, computers
spend most of their time displaying zooming multicolored
windowpanes or simulating an aquarium. These "screen saver"
programs, which compute nothing, whose only purpose is to
stir up pixels on the display screen, probably consume more
of the world's computational capacity than any other kind of software. Go into almost any office and you'll find machines busily saving their screens all night and all weekend.
Even when a computer is ostensibly in use, it is mostly
idle. Typing furiously, you might produce 10 keystrokes per
second; that's not much of a distraction for a processor
that can execute 100 million instructions in a second. Under these conditions the processor spends most of its time going around in a tight little loop, asking over and over, like a fidgety toddler, "What can I do now?"
This waste of computational machinery is not something we
need be ashamed of. The CPU cycles we fritter away today
will not be deducted from the legacy bequeathed to our
grandchildren. Still, every waste is also an opportunity,
and the cycles you have no use for may prove valuable to
The idea of scavenging unused cycles arose almost as soon as computers were linked by networks. A few early experiments with distributed computing, including a pair of programs called Creeper and Reaper, ran on the ARPAnet, the 1970s predecessor of today's Internet. Later, when the Xerox Palo Alto Research Center (PARC) installed the first Ethernet, a program cruised the network at night, commandeering idle computers for CPU-intensive tasks. This early cycle recycler was the creation of John F. Shoch and Jon A. Hupp, who called it a "worm," citing the inspiration of John Brunner's novel Shockwave Rider. (A colleague, noting the program's nocturnal habits, suggested the alternative name "vampire.") A later scavenger system called Condor, developed by Miron Livny and his colleagues at the University of Wisconsin at Madison, is now running at more than a dozen universities and other sites. Condor roams within clusters of Unix workstations, usually confined to a single laboratory or department.
Going out over the public Internet to scrounge cycles is
more difficult because the machines are more diverse and the network connecting them is more tenuous. Furthermore,
communicating with the machines is only part of the problem; making connections with the machines' owners is also harder. Nevertheless, Internet "metacomputing" has been going on for at least a decade. Here are some of the notable projects.
» Post Comment