Scientific IEEE 754 32-Bit Single-Precision Floating-Point Datasets
Click on the names of the datasets below to download them. These datasets are the single-precision counterparts to the double-precision datasets available here. By downloading one or more of the datasets, you agree:
- not to distribute any of them without acknowledging the source (i.e., this web page), and
- to acknowledge the source and to cite at least one of the following papers wherever you publish results derived from these datasets.
M. Burtscher and P. Ratanaworabhan. "High Throughput Compression of Double-Precision Floating-Point Data." 2007 Data Compression Conference, pp. 293-302. March 2007. [pdf]
M. Burtscher and P. Ratanaworabhan. "FPC: A High-Speed Compressor for Double-Precision Floating-Point Data." IEEE Transactions on Computers, Vol. 58, No. 1, pp. 18-31. January 2009. [pdf]
All datasets have been compressed with SPDP. To decompress them, follow the instructions on the SPDP webpage.
Observational data: These four datasets comprise measurements from scientific instruments.
- obs_error (22 MB): data values specifying brightness temperature errors of a weather satellite
- obs_info (5 MB): latitude and longitude information of the observation points of a weather satellite
- obs_spitzer (99 MB): data from the Spitzer Space Telescope showing a slight darkening as an extrasolar planet disappears behinds its star
- obs_temp (20 MB): data from a weather satellite denoting how much the observed temperature differs from the actual contiguous analysis temperature field
Numeric simulations: These four datasets are the result of numeric simulations.
- num_brain (52 MB): simulation of the velocity field of a human brain during a head impact
- num_comet (45 MB): simulation of the comet Shoemaker-Levy 9 entering Jupiter's atmosphere
- num_control (78 MB): control vector output between two minimization steps in weather-satellite data assimilation
- num_plasma (1 MB): simulated plasma temperature evolution of a wire array z-pinch experiment
Parallel messages: These five datasets contain the numeric messages sent by a node in a parallel system running NAS Parallel Benchmark (NPB) and ASCI Purple applications.
- msg_bt (90 MB): NPB computational fluid dynamics pseudo-application bt
- msg_lu (72 MB): NPB computational fluid dynamics pseudo-application lu
- msg_sp (96 MB): NPB computational fluid dynamics pseudo-application sp
- msg_sppm (20 MB): ASCI Purple solver sppm
- msg_sweep3d (14 MB): ASCI Purple solver sweep3d
We kindly thank the following people and institutions for providing the data from which the above datasets were distilled:
- Prof. Joseph Harrington of the Department of Physics at the University of Central Florida provided the datasets obs_spitzer and num_comet.
- Prof. David Hammer and Ms. Jiyeon Shin of the Laboratory of Plasma Studies at Cornell University provided num_plasma.
- Mr. Sami Saarinen of the European Centre for Medium-Range Weather Forecasts provided obs_temp, obs_error, obs_info, and num_control.
- We generated num_brain using a modified version of EULAG, a fluid code developed at the National Center for Atmospheric Research in Boulder, Colorado.
- Mr. Jian Ke ran the NPB and ASCI Purple benchmarks with 64 processes to capture the five message datasets.
This work is supported in part by the Department of Energy under Grant No. DE-FG02-06ER25722.