Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 17 additions & 0 deletions Docs/Intro and Specs/Supported compute modules.md
Original file line number Diff line number Diff line change
Expand Up @@ -157,4 +157,21 @@
<p>If you choose to purchase the module without the built-in eMMC, the SD card on the CM4 adapter board can be utilized, and the operating system can be booted from it. However, if you opt for the version with the eMMC extension, you will need to use the USB eMMC adapter tool provided by Pine64 to flash the operating system onto the module.</p>
<p class="wysiwyg-text-align-center"><img src="https://help.turingpi.com/hc/article_attachments/9170063145629" alt="pine64_3.png"></p>
<p class="wysiwyg-text-align-center"><img src="https://help.turingpi.com/hc/article_attachments/9169990889885" alt="pine64.png"></p>
<h2>BigTree-Tech CB1 Compute Module</h2>
<p class="wysiwyg-text-align-center"><img src="https://cdn.shopify.com/s/files/1/1619/4791/products/5_b06e0a40-0912-4ded-8a0e-4831c6c860a8_600x.jpg?v=1670830487" alt="bigtreetech-cb1.png"></p>
<p>The BigTree-Tech CB1 compute module was designed and marketed for 3D Printer control board applications
but functions as a general purpose CM4 form factor compute module. Details of its specs can be found in the
<a href="https://github.com/bigtreetech/CB1/blob/master/BIGTREETECH%20CB1%20User%20Manual.pdf" target="_new">product manual on their github</a>.
In short, the CB1 is a quad core, A53 SOC with either 512MB or 1GB of ram.</p>
<p>In testing, three 1GB CB1 modules were paired with a Raspberry Pi CM4 with 4GB of ram. A CB1 is unlikely to support
a control plane node for k3s or microk8s by itself, thus the CM4 served as the control plane. The CB1 modules had
<a href="https://github.com/bigtreetech/CB1/releases/tag/V2.3.2" target="_new">CB1_Debian_11_minimal from the BigTree-Tech Github</a>
installed via a 32GB microSD card. The CM4 ran <a href="https://downloads.raspberrypi.org/raspios_arm64/images/raspios_arm64-2023-02-22/2023-02-21-raspios-bullseye-arm64.img.xz" target="_new">2023-02-21-raspio-bullseye-arm64</a>.</p>
<p>The same configuration of microk8s and k3s were deployed via Ansible to all nodes with no special configuration required
on the Turing Pi 2 itself. Logs for the test are available
<a href="https://raw.githubusercontent.com/nuvious/HomelabKubernetes/003-Armbian-Support/docs/20230421-Test.log" target="_new">here</a>
(hostnames tpi21, tpi22, and tpi23 were CB1 modules) and the ansible playbooks used are available
<a href="https://github.com/nuvious/HomelabKubernetes/tree/8c7ba8be7b09c037c4bdb29f511ef858dce9f9c6">here at the linked commit</a>.
If used in a kubernetes implementation, k3s will offer less overhead than microk8s which should be considered given the limited
resources of a CB1 modules. All that said, the BigTree-Tech CB1 compute module has been tested and works on the Turing Pi 2.</p>
<p> </p>