Sebastopol, CA--O'Reilly's latest release,
Clusters by David HM Spector, defines a cluster in its simplest form
as "a bunch of computers tied together with a network, working on some large
problem that has been broken down into smaller pieces." The simplicity of
the definition belies the importance of the subject. The list of the world's
most powerful supercomputers now includes Beowulf machines: PCs running Linux,
clustered together into a single machine.
From scientific applications to transaction processing, clustering
technology provides an affordable, scalable computing solution. One of the
hottest topics in Linux today is the ability to cluster Linux machines to
mimic the performance of supercomputers (costing hundreds of thousands of
dollars) for a fraction of the cost.
"Computing problems always seem to meet or exceed available computing
resources. There is always a need for more processing power, more network
bandwidth, and greater I/O capabilities," says Spector. "This need for
speed has prompted a whole new class of machines: supercomputers. In fact,
supercomputer class machines-the construction of which is the topic of my
book-influence almost every aspect of modern life."
"The Linux operating system has given rise to a host of interesting and
unique computing possibilities that would have been impossible, and perhaps
even unthinkable, in a pre-Linux era," says Spector. Some examples that
Spector gives, include:
- A single cross-platform operating system
- An open source operating system whose entire code base is freely
available for research, commercial use, and improvement
- A system that has the potential to displace both Unix variants and
Windows NT as a server operating system
- A scaleable parallel processing system that allows performance comparable
to traditional supercomputers (IBM SP/2 or Cray), but at a fraction of the
"Since the ground-breaking work was done at NASA, Beowulf-style clusters
have become their own well-defined genre of high-performance computing
systems," explains Spector. "By October 1999, several of the systems in the
top 200 supercomputers in the world were made up of Linux cluster systems."
Spector's first Linux cluster, and ultimately his new book
Clusters, arose from his need for an affordable home supercomputer for
his research. Building it was a challenge. "I noticed that a lot of people talk
about it, but the documentation is really terse. There's no way for someone
to start who doesn't have all the knowledge already." Spector wrote up an
outline and realized it would make a great O'Reilly book. And, as it turned
out, O'Reilly was looking for just such a book.
Building Linux Clusters is a hands-on guide for people new to clusters.
"Like most O'Reilly books," said Spector, "it's to get people going. You
can build a cluster right out of the box: You pop the CD in, follow the
instructions, and in ten or twenty minutes you have a fully functional
Spector likes to say that the book tells a story. And indeed, the first
chapter is a brief, lively history of the evolution of supercomputers and,
eventually, of clusters, including Don Becker's first Beowulf cluster.
Built in 1994 to meet NASA's need for supercomputer performance on a
shoestring budget, it used Linux and off-the-shelf hardware, saving NASA a
couple of million bucks, according to Spector.
Spector thinks that the most exciting future applications of clusters are
the least possible to predict. That's because learning to work with
parallel computing changes the way people look at problems and opens new
possibilities for applications. "People have to step out of the mindset of
being serial and think about how to do everything at the same time. And
that's a hard transition, but once you do that, suddenly new vistas open to
Scaling Linux for Scientific and Enterprise Applications
By David HM Spector
1st Edition, August 2000
1-56592-625-0, 352 pages, $44.95, Includes CD-ROM