Changes between Version 20 and Version 21 of Grass8Planning


Ignore:
Timestamp:
Nov 25, 2017, 10:02:51 AM (6 years ago)
Author:
wenzeslaus
Comment:

virtual maps for tile support

Legend:

Unmodified
Added
Removed
Modified
  • Grass8Planning

    v20 v21  
    2626 * save more raster metadata like number of non-null cells, mean and stddev (see GDAL)
    2727 * MASK: add support for writing raster data (not only reading)
    28 
    29 Discussions at Aug 2017:
    30  * add **tile support** for better large map support (Sentinel, global data, ...), supporting massive parallel computations
     28 * add **tile support** for better large map support (Sentinel, global data, ...), supporting massive parallel computations (based on discussions from Aug 2017)
     29  * it must be clear why this would better than GDAL vrt combined with r.external
    3130  * tiles could then be even stored on different nodes for speed optimization
    32   * e.g. develop a new virtual raster mapset "VRT" (special like PERMANENT)
    33    * reading:
    34     * or add tile support deeply into raster lib (Rast_get_row()
    35     * use name scheme? make use of segment library
    36     * problem: due to row compression always whole row is read even if computation region is smaller
    37    * writing:
    38     * it is more complex
     31  * storage implementation:
     32   * develop a new virtual raster mapset "VRT" (special like PERMANENT)
     33   * virtual map: combination between current groups/stds and external maps - map metadata link the other maps, possibly in other mapset (Vaclav: I suggest this (vrt map) rather than vrt mapset or low level (i.e. new format) tiling)
     34    * something like the segment library opens the appropriate "standard" raster maps
     35    * different storage nodes possible if maps are different mapsets which are on diff nodes
     36  * reading:
     37   * [the above options or] add tile support deeply into raster lib (Rast_get_row())
     38   * use name scheme? make use of segment library
     39   * problem: due to row compression always whole row is read even if computation region is smaller
     40  * writing:
     41   * it is more complex
    3942
    4043=== Imagery library ===