Counted among the most dramatic predictions of general relativity are black holes and gravitational waves. Scientists are now poised to probe general relativity, black holes and the early universe using the Laser Interferometer Gravitational-wave Observatory (LIGO) as an astronomical tool. The success of LIGO's long-term observing campaign is predicated on continuing timely analysis of the observational data. To achieve this non-trivial computational task, LIGO has successfully adopted the grid computing model and has been at the forefront of its application for data-intensive observational science. This award supports the LIGO Data Grid, a distributed computational facility developed specifically to meet the significant computational needs of gravitational-wave astronomy. It provides personnel to support, enhance and deploy software and grid-services on the LIGO Data Grid. It addresses the needs of the community of gravitational-wave physicists involved in LIGO data analysis, and will have significant impact in that area. The LIGO Data Grid also tests grid concepts at both the high level of grand ideas and at the detailed level of tools in everyday use for a large scientific experiment. This experience will have broad relevance for other grid and information technology initiatives. Finally, this award will provide training for students, scientists and computer specialists in the use of emerging cyber-technology which has applications outside LIGO's endeavor.
The Laser Interferometer Gravitational-wave Observatory (LIGO) houses the most sensitive gravitational-wave detectors ever built. They were designed to search for minute ripples in spacetime curvature that travel to us from colliding black holes, neutron stars, and the early universe. While there is strong indirect evidence for gravitational waves from the changing orbit of the binary pulsar PSR1913+16, gravitational waves have never been directly measured. Special algorithms have been designed to process data from the LIGO detectors and to identify sufficiently strong gravitational-wave signals. The success of LIGO's observing campaigns was predicated on continuing timely analysis of the observational data. This award supported the LIGO Data Grid, a distributed computational facility developed specifically to meet the significant computational needs of gravitational-wave astronomy. It provided personnel to support, enhance and deploy software and services on the LIGO Data Grid. Scientists operated the initial LIGO detectors for two long science runs from November 2005 to November 2007 and from July 2009 to October 2010 during which they recorded about 0.5 PetaBytes of data. No significant gravitational-wave signals were observed in the data from these two science runs. Nevertheless, these searches for gravitational waves have identified upper limits on the strength and rate of gravitational waves from the big bang, from collisions of black holes and neutron stars, and from more exotic sources including cosmic strings. And these searches would not have been possible without the software and computational resources provided by the LIGO Data Grid. In addition to support of the initial LIGO science runs, this award supported development of the computing model for Advanced LIGO later this decade: the advanced detectors are expected to reveal the long sought gravitational-wave signals. The LIGO Data Grid has also provided a testbed for computational concepts in big-data at both the high level of grand ideas and at the detailed level of tools in everyday use for a large scientific experiment. Finally, this award provided training for students, scientists and computer specialists in the use of emerging cyber-technology which has applications outside LIGO's endeavor. A number of students and staff, who received their first introductions to big data as part of this project, have moved into the industrial sector as systems administrators, analysts, and developers.
