Benchmarks of file and image generation
vinceskahan edited this page Feb 27, 2024
·
30 revisions
See https://github.com/weewx/weewx-benchmark. Follow the directions in the readme for how to run the benchmark.
When you're done, add your results to the table below. For architecture on
Debian systems, please run dpkg --print-architecture
, rather than relying on
uname -a
. If that does not reveal anything useful, then try cat /proc/cpuinfo
.
Hardware | WeeWX | Python | Files (21) | Images (68) | Footnotes |
---|---|---|---|---|---|
Mac Mini, M1 2020 (Sonoma 14.0) | 5.0.0b13 | 3.10.9 | 0.60s | 1.06s | |
MacBook Air, M1 2020 | 4.10.2 | 3.7.16 | 0.69s | 0.78s | |
MacBook Air, M1 2020 | 5.0.2 | 3.7.16 | 0.69s | 0.93s | |
NUC i7, 11th gen | 5.0.0b13 | 3.7.10 | 0.89s | 1.14s | |
Xen VM running on i5-9600K 3.7GHz | 5.0.0rc1-1 | 3.11.2 | 1.44s | 1.51s | |
RPI 5 Model B Rev 1.0 (4GB RaspiOS lite arm64 6.1.0-rpi6-rpi-2712 ) |
5.0.0b15 | 3.11.2 | 1.63s | 2.03s | |
Intel 4th gen NUC D54250WYKH1 (mid-2014 i5-4250U, Ubuntu 24.04 LTS) |
5.0.2 | 3.12.2 | 2.19s | 2.31s | |
Macbook Air 5,2 (mid-2012 IvyBridge i5, Ubuntu 22.04.3 LTS) |
5.0.0b13 | 3.10.12 | 2.41s | 2.63s | |
RPi 4 Model B Rev 1.1 (4GB, RaspiOS lite arm64 6.1.0-rpi4-rpi-v8) |
5.0.0b13 | 3.11.2 | 6.24s | 6.39s | |
RPi 4 Model B Rev 1.1 (4GB, RaspiOS lite armhf 6.1.21-v8+) |
5.0.0b13 | 3.9.2 | 6.36s | 6.56s | |
RPi 3 Model B Plus Rev 1.3 (RaspiOS lite armhf 6.1.21-v8+) |
5.0.0b13 | 3.9.2 | 13.06s | 14.07s | |
inovato quadra Allwinner H6 ARMv8 Cortex-A53 @ 1.7GHz |
5.0.0b13 | 3.9.2 | 14.03s | 11.87s | (1) |
MacBook Pro, M1 Max, 2021 (Sonoma 14.3) in a Docker Container | 5.0.0-1 | 3.7.3 | 18.43s | 19.17s | |
RPi 2 (32-bit) | 5.0.0b13 | 3.9.2 | 24.27s | 25.95s | |
Rpi Zero W (32-bit) | 5.0.0b13 | 3.9.2 | 53.97s | 57.79s |
(1) Using the Python version of Cheetah's NameMapper, which can significantly slow file generation.