Skip to main content
. Author manuscript; available in PMC: 2015 Nov 1.
Published in final edited form as: Ecol Inform. 2014 Nov 1;24:11–16. doi: 10.1016/j.ecoinf.2014.06.004

Table 1.

Sample running times for Aardwolf queries. Machine specifications: 2.6 GHz Intel dual core i5, 8GB 1600 MHz DDR RAM, 500 GB flash drive. Query running times are given in milliseconds (ms), seconds (s) and minutes (m). The database sizes for N photographs were N=1000: 1MB, N=100,000: 30MB, N=1,000,000: 300MB, N=10,000,000: 3GB.

Query Query running time per N
Notes
Definition of N N=1000 N=100000 N=1M N=10M
List cameras Number of cameras 1ms 30ms 350ms 3s 500ms In practice number of cameras < 100
List folders Number of folders 1ms 30ms 250ms 2s 500ms In practice, number of folders < 1000
List folders after applying filters Number of photographs 1ms 20ms 3s 300ms 8s 500ms Base case of listing 1 folder with N photographs
List folders after applying filters Number of folders and number of photographs 1ms 200ms 8s 1m 10s Query times increase with number of folders and number of photographs
List photographs in a folder Number of photographs in a single folder 1ms 30ms 250ms 3s 200ms In practice, folders contain <10,000 photographs in Aardwolf because each folder corresponds to a single download event from a camera storage card
List photographs in a folder after applying filters Number of photographs in a single folder 1ms 150ms 1s 450ms 19s 500ms
Search tags in photographs Number of photographs in project 1ms 250ms 2s 300ms 32s 500ms
View data Number of photographs in project 15ms 1s 400ms 15s 500ms 3m 20s
Export data to a CSV file Number of photographs in project 1s 3s 500ms 30s 500ms 5m 35s This operation depends primarily on write speed for target device of the CSV file. These numbers are from a machine with a flash drive, which is typically several times faster than a traditional hard drive.