Maria Map Maker raster import benchmark: Difference between revisions
Jump to navigation
Jump to search
(Created page with "The following benchmarks was perfomed to indicate the decrease in processing time for raster import in Maria Map Maker. We have used three different raster datasets on three different versions of M3. == System == * Windows Server 2019 * Intel Core i9-10900 (10 cores, 20 logical processors) * 128 GB RAM * 2 TB NVMe disk == Input datasets == {| class="wikitable" |+ !Name !Format !No. files !File size !Pixel size |- |N2000 |GeoTIFF |1 |48 MB |80 m |- |ECW Innlandet |EC...") Â |
(→â€)  |
||
| (5 intermediate revisions by the same user not shown) | |||
| Line 7: | Line 7: | ||
* 128 GB RAM | * 128 GB RAM | ||
* 2 TB NVMe disk | * 2 TB NVMe disk | ||
* On Maria Map Maker v4.4, the new parameter ''CPU use'' was set to use 15 logical processors | |||
== Input datasets == | == Input datasets == | ||
| Line 37: | Line 38: | ||
== Results == | == Results == | ||
[[File: | The results show that there is a small overall decrease in processing time for single-file datasets. For very small datasets there is a few seconds increase in time between v4.3 and v4.4 due to a completely different pipeline. For large datasets consisting of a large number of files, the decrease in processing time between 4.2 and 4.4 is around 900%. | ||
[[File:M3 rasterimport benchmark ECWInnlandet.png| | Â | ||
[[File:M3 rasterimport benchmark N50.png| | [[File:M3_rasterimport_benchmark_N2000.png|frameless|550x550px]] | ||
 | |||
 | |||
[[File:M3 rasterimport benchmark ECWInnlandet.png|550x550px|border|frameless]] | |||
 | |||
 | |||
[[File:M3 rasterimport benchmark N50.png|550x550px|border|frameless]] | |||
Latest revision as of 13:36, 18 October 2022
The following benchmarks was perfomed to indicate the decrease in processing time for raster import in Maria Map Maker. We have used three different raster datasets on three different versions of M3.
System
- Windows Server 2019
- Intel Core i9-10900 (10 cores, 20 logical processors)
- 128 GB RAM
- 2 TB NVMe disk
- On Maria Map Maker v4.4, the new parameter CPU use was set to use 15 logical processors
Input datasets
| Name | Format | No. files | File size | Pixel size |
|---|---|---|---|---|
| N2000 | GeoTIFF | 1 | 48 MB | 80 m |
| ECW Innlandet | ECW | 1 | 1440 MB | 0.25 m |
| N50 | GeoTIFF | 1335 | 16.6 GB | 2.5 m |
Results
The results show that there is a small overall decrease in processing time for single-file datasets. For very small datasets there is a few seconds increase in time between v4.3 and v4.4 due to a completely different pipeline. For large datasets consisting of a large number of files, the decrease in processing time between 4.2 and 4.4 is around 900%.
