We have been monitoring this undertaking since mid 2021, and the time has been nicely spent. Ivan Kuleshov’s Compute Blade is a skinny PCB that packs a plethora of storage choices on your Raspberry Pi Compute Module 4 (or suitable). Kuleshov’s kickstarter has smashed its $522,209 funding objective, reaching $673,365 on the time of writing.
The Compute Blade is a rack-mountable provider board for the Raspberry Pi Compute Module 4, designed for high-density clusters. The PCB is filled with options, however your eye can be drawn to the crimson anodized aluminum heatsink which inserts over the Compute Module 4 (or suitable), offering a passive means to maintain the Pi cool. This might show helpful, must you want to overclock.
Row 0 – Cell 0 | Uptime Compute Blade Fundamental | Uptime Compute Blade TPM | Uptime Compute Blade Dev |
Raspberry Pi CM4 Connector | Sure | Sure | Sure |
M.2 (M Key as much as 22110 NVMe drives) | Sure | Sure | Sure |
Gigabit Ethernet with PoE | Sure | Sure | Sure |
UART | Sure | Sure | Sure |
Compute Blade Headers | Sure | Sure | Sure |
Stealth Mode | Sure | Sure | Sure |
LEDs | Sure | Sure | Sure |
USB-A port | No | Sure | Sure |
TPM | No | Sure | Sure |
HDMI | No | No | Sure |
USB-C (Bootloader replace) | No | No | Sure |
Micro SD card slot | No | No | Sure |
Wi-Fi, BT, nRPIBOOT | Locked | Locked | Switchable |
The PCB measures 27 x 4 x 1.6 cm, and Kuleshov has optimized the board for a lot of completely different options. Going from the Ethernet port on the left, now we have Gigabit Ethernet supporting Energy over Ethernet (PoE), on our dev unit there may be an HDMI port. We then attain the anodized heatsink, secured utilizing T7 hex screws. After that. there’s a USB-A port, adopted by an expansive M.2 slot. This slot helps NVMe drives from 2230 to 22110, or it may be used with different modules reminiscent of Google’s Coral TPU.
In case you want entry to the GPIO, you’ve some GPIO pins at your disposal. We do not get the complete 40 pins, which suggests we will not connect HATs, however we do have entry to a small variety of GPIO pins. If we have to join sensors, then the I2C pins are uncovered and can be utilized with parts supporting the communication protocol, together with Stemma QT parts.
The Compute Blade’s energy is available in numbers, extra particularly “clusters”. Given the small measurement and blade design of the models, they’ll simply slide right into a blade server and so long as you have acquired loads of Raspberry Pi’s, you may have a robust Arm computing cluster.
With costs ranging from $65 for a Compute Blade Fundamental, the model now we have on the bench is the $107 Dev model, which has all of the bells and whistles. In case you like what you see, then head over to the kickstarter web page to make your pledge. We’ll be chatting reside with Kuleshov on February 14 at 2pm ET / 7pm GMT on our weekly Raspberry Pi present, The Pi Forged.
Keep in mind that crowdfunding a undertaking will not be a assure of receiving a completed product. Backing a crowdfunded undertaking is akin to an funding; you imagine within the undertaking and need it to succeed. You aren’t buying a retail product.