Despite all the reports of the rapidly shrinking globe—with modern technology like the Internet, wireless handheld computers, and satellite global positioning systems—the universe seems to get lonelier every day. The search for intelligent life in the universe continues, although nothing has been found…yet. That "yet" is what keeps people looking and actually, more specifically, listening. In fact, thanks to recent developments the search will soon be stepped up considerably.
The Search for Extraterrestrial Intelligence (SETI) program
, derided in some scientific circles as a waste of time, received a good measure of publicity with the success of Cal Berkeley's groundbreaking SETI@Home project.
Too Much Data
SETI@Home, just one part of the larger Project Phoenix at UC Berkeley, is an experiment in shared computer resources. It would take way too long to analyze the massive amounts of data that SETI collects, but in 1994 David Gedye and Craig Kasnoff, two SETI vets, had a conversation at a Christmas party that changed SETI forever. They figured out that instead of trying somehow to get outrageous sums of money for a massive super computer to crunch all the data they collected, they could use smaller computers. A smaller computer would normally take much longer than a large computer, but they asked: What if you had thousands, even millions, of smaller computers doing the work? Thanks to the Internet, it was a possibility. That was how SETI@Home was born.
The Power of the Internet
The project really got rolling when the website was launched in May 1999. It was originally slated to end in two years, but thanks to its wildly surprising popularity and additional funding, it has been prolonged indefinitely. By now, more than 3 million volunteers in over 220 countries have gone to the SETI@Home website and downloaded the software in order to use their computers for the cause.
A teraflop is a measure of a computer's speed and can be expressed as:
A trillion floating point operations per second
10 to the 12th power floating-point operations per second
2 to the 40th power flops
Scientists have begun to envision computers operating at petaflop speeds.
This is how it works. The software downloads a batch of data from SETI, then when the computer is not being used, the program kicks in like a screensaver program would and starts analyzing the information. Once the computer has analyzed a batch, it sends the data back to SETI and receives new data, starting the process again.
The power and cost efficiency of the giant network of personal computers is boggling. Consider this: The most powerful computer, IBM's ASCI White, is rated at 12 >teraflops and costs $110 million. The computational power of SETI@Home runs at about 15 teraflops and so far has cost just $500,000.
High Costs Lead to Innovations
The Project Phoenix observations are currently being made using the 1,000-foot radio telescope at Arecibo
, Puerto Rico, and the scientists have gone through a large portion of the stars on the Phoenix hit list, listening to two billion channels for each star targeted. But the dish at Arecibo is used for SETI-related purposes only a fraction of the time.
The SETI project will soon have its own SETI-dedicated telescope to broaden the search. The idea for an all-SETI telescope is not new, but earlier efforts always failed due to the high costs. The new Allen Telescope Array(ATA), to be built in northern California, will differ in practice, appearance, and cost from optical and radio telescopes currently in use. It will be constructed using hundreds of mass-produced small dishes. The telescope will use new technologies along with the large amounts of affordable computer processing that the SETI@Home project has unearthed. By doing so, it will be possible for the Allen Telescope Array to examine up to a dozen SETI target stars simultaneously and be sensitive to signals over a very wide range of frequencies.
New Observatory in Northern California
Under current plans, the Allen Telescope Array will be developed in two phases. The first phase began last year with the development of the prototype unveiled in April, and will culminate with a second, larger prototype in early 2003, one that will actually get started on SETI and radio astronomy research. At that point, with all the new technologies proven, a second-stage technical and funding review will occur.
The ATA's five-meter antennas are mass-produced and extremely cost-effective. Named for benefactor and Microsoft co-founder Paul Allen, the array will be located at the Hat Creek Observatory, 290 miles northeast of San Francisco on a site operated by UC Berkeley's Radio Astronomy Laboratory. The Hat Creek Observatory is located in an area that is "radio quiet," thereby reducing the level of interfering signals from man-made sources.