Developer Builds Cluster with 9 Raspberry Pi Pico Boards
Given that the Raspberry Pi Pico and the RP2040 chip that powers it are microcontrollers, it’s amazing how much computing power developers were able to squeeze out of them. Running at 133 MHz and with 256K of his SRAM, the RP2040 chip is used in everything from handheld game consoles. (opens in new tab) For web servers and plant monitoring systems. Of course, you can also play Doom.
UK-based developer Derek Woodroffe currently operates the following sites and Twitter accounts: extreme electronics (opens in new tab)built a method to combine nine different Raspberry Pi Picos into a distributed computing cluster. On his Twitter he posted a video of the cluster performing the Mandelbrot calculation and using an I2C connection he outputs as a fractal image on a 240 x 320 display (ili9341).
9 picos , 1 display… still some issues with missing data. I suspect more of a race condition in git. https://t.co/WZ7SRE3ibEpic.twitter.com/wocjRiXod1April 7, 2023
Woodroffe has posted code and instructions to reproduce a project he calls PicoCray. on his github page (opens in new tab)He wrote that GPIO 10, 11, 12, 13 and GND must be connected from one board to another in order to pair the Picos. Each Pico must be powered independently via USB or a 3.3 volt pin. He also states that on the last processor in the group he should place 4.7K ohm resistors on GPIO 10 and 11.
The code is identical for all nodes in the cluster and if one is to be used as the controller, a wire from GPIO 22 to ground must be connected. The github page has more details.
Woodroffe says that at this point, it’s more of a proof of concept than a powerful multiprocessor system. He wrote, “In many uses it’s actually slower than running on a single Pico.” However, the idea is solid and we hope more developers will take advantage of it.
To be fair, this isn’t the first time multiple RP2040 chips have been used together to boost computing power. Recently, Pimoroni announced that he is working on a new DV stick with one He RP2040 for video and another He RP2040 for other tasks. And when it comes to regular Raspberry Pi computers, there are many cluster options. One mega cluster contained 100 Pi (opens in new tab).