rick@hanauma (Richard Ottolini) (05/11/89)
It was my impression that most of the ten terrabytes of the USGS database would be consumed by the topographic contour lines rather than the cultural overlays. Ten gigabytes for each of the dozen or so overlays sounds consistant. Just wait until airplane or satellite image data is incorporated as the USGS plans to do and military mapping agencies are already attempting! 3000 EW km x 10 pixels / km x 1500 NS km x 10 x 10 spectral bands x 1 byte / band equals 450 GB.
eric@batcomputer.tn.cornell.edu (Eric Fielding) (05/15/89)
In a recent article rick@hanauma (Richard Ottolini) wrote: >.... Just wait until airplane or satellite image data is incorporated >as the USGS plans to do and military mapping agencies are already attempting! >3000 EW km x 10 pixels / km x 1500 NS km x 10 x 10 spectral bands x 1 byte / band >equals 450 GB. (this does not multiply out right on my calculator. I get 4.5 GB.) 10 pixels per km is pretty low resolution. The Landsat Thematic Mapper data (with complete coverage of the US several times over) has 30 m pixels and seven bands: 3000 km x 30 p/km x 1500 km x 30 x 7 bands = 28 GB. There is the SPOT imagery 10 m resolution x 1 spectral band = 450 GB I believe this is available for the full US several times over, also. Then there is the Eos (Earth Observing System) NASA is planning to launch in the mid-90s. It will have an instrument call HIRIS with 30 m pixels and up to 128 bands (limited by the rate at which they can beam it down), with a 30 km wide swath. This should cover the US several times a year. They are now trying to figure out how to deal with this data volume. ++Eric Fielding fielding@geology.tn.cornell.edu (BTW, I analyzed about 2.5 GB of Landsat Thematic Mapper data for my PhD dissertation. The new digital videotape storage devices make this pretty easy.)
littauer@uts.amdahl.com (Tom Littauer) (05/18/89)
In article <7949@batcomputer.tn.cornell.edu> fielding@geology.tn.cornell.edu writes: > >10 pixels per km is pretty low resolution. The Landsat Thematic Mapper data >(with complete coverage of the US several times over) has 30 m pixels and >seven bands: 3000 km x 30 p/km x 1500 km x 30 x 7 bands = 28 GB. > >There is the SPOT imagery 10 m resolution x 1 spectral band = 450 GB >I believe this is available for the full US several times over, also. Perhaps we miscalculated at the time, but we figured it'd take about 100 GB to do "your town from space" instead of "road map of your town". Cost of the data was all that kept us from doing it... > >Then there is the Eos (Earth Observing System) NASA is planning to launch in >the mid-90s. It will have an instrument call HIRIS with 30 m pixels and >up to 128 bands (limited by the rate at which they can beam it down), with >a 30 km wide swath. This should cover the US several times a year. They are >now trying to figure out how to deal with this data volume. > Ahem. :-) > ++Eric Fielding >fielding@geology.tn.cornell.edu > >(BTW, I analyzed about 2.5 GB of Landsat Thematic Mapper data for my PhD >dissertation. The new digital videotape storage devices make this pretty >easy.) 2.5 GB is the smallest amount we sell... it'd have been even easier on 17ms. disk, yes? Sorry for the blatant commercialism, but you guys need to rethink what BIG is. Yes, I know it ain't cheap - but it's been shipping for a long time and works; and it makes some problems solvable that are out of reach on small systems. -- UUCP: littauer@amdahl.amdahl.com or: {sun,decwrl,hplabs,pyramid,ames,uunet}!amdahl!littauer DDD: (408) 737-5056 USPS: Amdahl Corp. M/S 278, 1250 E. Arques Av, Sunnyvale, CA 94086 I'll tell you when I'm giving you the party line. The rest of the time it's my very own ravings (accept no substitutes).