The volunteer computing project, Einstein@Home, is helping science lovers around the world advance scientific knowledge about gamma-ray pulsars, gravitational waves, and a galactic global positioning system. All that volunteers have to do to partake in this cutting-edge research is donate their idle computing power.
As with most volunteer computing projects, Einstein@Home accesses the computing power it needs via an open-source application called BOINC – the Berkeley Open Infrastructure for Network Computing – which was originally created for the SETI@Home project. To join, volunteers download and install BOINC and select the projects they wish to participate in. BOINC runs in the background, providing CPU and GPU cycles on the volunteers’ computers that would otherwise go unused.
Einstein@Home was launched in 2005 to help analyse data from the Laser Interferometer Gravitational Wave Observatory (LIGO). LIGO is searching for direct evidence of the existence of gravitational waves – ripples in the fabric of the space-time continuum predicted by Albert Einstein’s theory of General Relativity.
Today, Einstein@Home is in search of much more than gravitational waves. In addition to LIGO data, Einstein@Home is analyzing publicly-available data from the Fermi Gamma-ray Space Telescope, the Arecibo Observatory, and the Parkes Radio Telescope. In the past six months, more than a dozen pulsars have been discovered by the volunteer computing project. Pulsars are young neutron stars – city-sized spherical remnants of a star’s supernova that produce focused beams of radio waves or gamma-ray photons. As neutron stars age, their rotation slows until they are no longer classified as pulsars.
“Some neutron stars spin up to 700 times per second around their axis, some of them faster than your average kitchen blender or Formula One racing engine,” said Benjamin Knispel, a researcher at the Max Planck Institute for Gravitational Physics (Albert Einstein Institute) in Hanover, Germany, who collaborates on the project.
Pulsars are in turn being used as a type of galactic Global Positioning System to search for gravitational waves.
Although the volunteer computing project has been analysing science-quality data since June 2006, so far no gravitational waves have been found.
“These methods were originally designed to identify gravitational waves from neutron stars, the primary goal of our project,” said Bruce Allen, leader of the Einstein@Home collaboration. “We adapted it for other types of data because we wanted to motivate our volunteers who have been running Einstein@Home for a while, and to help discover more stars.”
That’s why, in 2009, the Einstein@Home application began to analyse other sources of data to search for pulsars. To do so, the collaboration used a method called a ‘blind search,’ which looks for stars spinning at any rate anywhere in the sky.
“This method was able to search for a wider range of frequencies and a larger area of the sky,” Allen said. “We identified signatures of pulsars that previous searches of the same data missed. These methods are an optimal method to finding new pulsar.”
Although volunteers are far more likely to discover a new pulsar than to detect a gravitational wave, the chances of finding a previously unknown star are the same as winning the lottery.
“The chance of finding a new pulsar is one in 10 million. In less than two years 27 have been found so far. The more you analyze Einstein@Home data, the more likely you will find a new star,” Allen said.
The first Einstein@Home ‘lottery winners’ were Daniel Gebhardt from Germany, and Chris and Helen Colvin from the US. They discovered a ‘disrupted recycled pulsar’ in 2010.
“When a new star is found and verified, I send out a personal email,” said Allen. “I sent a number of personal emails to the Colvins but they didn’t respond. In the end, I had to send a signed-for-letter by courier to make sure they got the message.”
The second discovery of a binary pulsar was partly made in February 2011 by Vitaliy V. Shiryaev, from Russia. He also happens to be chief of mathematics in Russia’s largest rail cargo operator, JSC Freight One.
“His company has a computing cluster that typically tries to solve the travelling salesman problem which, in their case, is finding the shortest possible route to get their goods to a destination,” Allen said. “Einstein@Home is run as a bottom feeder [a final task run when a computer is idle] on their system.”
Allen’s passion for the search for gravitational waves is steadfast. This elusive phenomenon could already be hiding in the terabytes of data that Einstein@Home receives.
When asked if volunteers could improve the blind search algorithm to find waves within the data, Allen sat back in his chair and said, “that’s a good question. In fact, in 2006, one of our volunteers, a Hungarian named Akos Fekete, contacted us with suggestions of how to improve our code. It turned out that Fekete is a programmer who specializes in writing assembly language.”
Fekete’s suggestions sped up some parts of their search code by 800%.
“Akos took the binary of Einstein@Home search code. He looked at the assembler code and figured out where it was spending most of its time. Then, he manually improved the assembler code to do the same calculations more efficiently, thereby speeding up the binary,” Knispel said.
Today, most of Einstein@Home’s source code is open for anyone to tinker with.
Paul Ray, an astrophysicist at the US Naval Research Laboratory, is particularly interested in studying the timing of gamma-ray emitting pulsars. Ray, along with a number of his colleagues with the Fermi Gamma-ray Space Telescope as well as other projects, believes that the timing of pulsars may offer an alternate way to search for gravitational waves. This is due to the unique regular signals emitted by pulsars, which are analogous to the ticking of a clock.
One project pursuing this line of inquiry is the International Pulsar Timing Array project, which aims detect gravitational waves by measuring signals from approximately 30 pulsars.
“International groups of astronomers are carefully timing the best pulsar clocks in hopes of seeing disturbances in the clock rate as gravitational waves from distant galaxies sweep over the Earth, stretching and squeezing space-time,” Ray said. “They are in a race with the ground-based laser interferometer experiments like LIGO and Virgo.”
The more pulsars found by Einstein@Home volunteers, the more understanding astronomers can gain about these stellar phenomena. This research can also help in future human space exploration. One day, pulsars could even be used as a GPS for space. As the most accurate natural clocks in the Universe, spread across the sky, they are the perfect natural analog to GPS.
“In principle, a spacecraft could use them as navigation beacons that would provide accurate time and position information, even very far from Earth, where GPS is not available,” Ray said. “A spacecraft heading to the outer solar system, or beyond, into interstellar space could use observations of pulsars to determine its location along the way.”
Currently, a number of new X-ray timing experiments are being decided upon to further this research.
NASA is assessing an X-ray timing mission called NICER. Ray said it has a roughly 50% chance of being selected for development next year.
Meanwhile, the European Space Agency will select one of three large missions (L class), one of which is called ATHENA.
“If they select Athena, then we will have a large X-ray spectroscopy and timing mission in development for launch early next decade,” Ray said.
ESA is also looking at four or five medium missions (M class), one of which is called LOFT, dedicated to X-ray timing of neutron stars and black holes. ESA will decide which mission to continue developing in the spring of 2013.
“In the event that LOFT is not selected,” Ray said. “A group of scientists led by myself and others, are preparing to propose a large area X-ray timing mission called AXTAR to the NASA Explorers Program.”