Using Docker on Banana Pi

In 2014 I watched new single-board-computers enter the market. I am very fond of the Raspberry Pi project and I am using RPi for quite some time now as you can see from my previous posts. Now I am curious how it compares to Banana Pi. I am currently looking for a base station for my home automation project. I wouldn't call this a server since this term is in most cases used for different hardware. In this context reliability, cost of operation running 24/7, ease of use are the most important factors for me. I am going to write a series of blog posts on benchmarks of the different models in this context. So stay tuned!

Banana Pi is another interesting single-board-computer in the 45$ price range from SinoVoip. It has a Dual 1GHz ARM Cortex CPU. What really distinguishes the Banana Pi board from the competition is 1GB SDRAM and 10/100/1000 Ethernet.

I downloaded and unpacked the Archlinux for BananaPi image (see resources). Then I put a sdcard into my laptop (Ubuntu) and transferred the image:

sudo dd if=./ArchLinux_For_BananaPi_v1412.img of=/dev/mmcblk0

Then I booted the BaPi from that sdcard and initiated sync and update:

$ pacman -Syyu
$ pacman-db-upgrade

An optional step is to configure a suitable hostname for your new BBB based docker machine. I choose "kujira03" for that.

$ hostnamectl set-hostname kujira03

And now the installation of the docker package itself:

$ pacman -S docker systemd

Enable docker to run as a service:

$ systemctl enable docker.service

Done! Now we will reboot the BaPi and check if everything worked as expected:

$ shutdown -r now

Check that the docker service is running:

$ ifconfig
...
docker0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
    inet 172.17.42.1  netmask 255.255.0.0  broadcast 0.0.0.0
    inet6 fe80::bd01:3011:3b12:b866  prefixlen 64  scopeid 0x20<link>
    inet6 fe80::5484:7aff:fefe:9799  prefixlen 64  scopeid 0x20<link>
    ether 56:84:7a:fe:97:99  txqueuelen 0  (Ethernet)
    RX packets 5  bytes 308 (308.0 B)
    RX errors 0  dropped 0  overruns 0  frame 0
    TX packets 45  bytes 6171 (6.0 KiB)
    TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
...

I am now able to use arm specific docker containers (as a convention they are usually prefixed with "armhf-"). In this example I start my "armhf-bench" container from docker hub and use it.

$ docker run -t -i finklabs/armhf-bench
107.8239s
879.7Kb/sec
578.979 MiB/s

That's it. Here my benchmark results for the Banana Pi:

* for the CPU measurement: lower is better

If you are interested in this work let me know.

Best, Mark