Sun Microsystems through its Academic Equipment Grant (AEG) program awarded the Geodesy Laboratory and the PANGA Data Analysis facility an Sun Enterprise 3500 server and a fibre channel disk array. This competitive award has significantly enhanced the data processing facility at the Laboratory. The Enterprise 3500 is currently configured with two 336 MHz Superscalar SPARC processors and 1 GB of memory. The computer is scalable up to eight processors using Sun's high-speed GigaplaneTM system and eight GB of memory. The server comes with 90 GB of internal disk storage. The system includes an UltraSCSI and fibre channel disk array that provides 109.2 GB external disk storage, two Creator3D graphics I/O boards, and a 20-40 GB 8mm Tape Drive. The system is powered by an UPS (Uninterrupted Power System).
We are using the Enterprise 3500 for calculating daily GPS solutions for our plate deformation and earthquake hazard investigations. We currently calculate about 2,000 GPS solutions per 24 hour day from 1991 to present, incorporating the PANGA (Pacific Northwest Geodetic Array) data and the global GPS array. This includes all data from BARD (Bay Area Regional Deformational Network) and SIGN (southern Californian Geodetic Array). We are processing approximately 75,000 station days/month, an eight-fold increase over 1998 laboratory productivity. This allows us to retroactively process global GPS data back to 1991.
The new Enterprise server has substantially enhanced our data processing and analysis. Since the machine has come on line in April, 1999, we have processed 300,000 daily station positions. This compares to 21,000 over the previous two years. We are now routinely processing 220 globally distributed GPS sites including the 30 sites that are formally part of the Pacific Northwest Geodetic Array. We have reprocessed all the available data back to 1995.
The scientific return on this enhanced capability is tremendous. The addition of a robust network of global sites poises us to monitor and model the regional data with great strength. We can use internally consistent methods for tracking how the Pacific Northwest is deforming relative to North America or the subducting sea floor that drives seismic risk. We can better understand how atmosphere, ocean loading and earth tides affect our data analysis. With this understanding we can contribute much more to constraints on seismic risk.
