Coral M2 Datasheet
Coral M2 Datasheet
Coral M2 Datasheet
2 Accelerator Datasheet
Version 1.0
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Table of contents
Features
Overview
Requirements
Specifications
Dimensions
A+E key dimensions
B+M key dimensions
Power specifications
Thermal limit and operating frequency
Connector pinout
A+E key pinout
B+M key pinout
Software and operation
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Features
Google Edge TPU ML accelerator
Photo shows A+E key form factor with shield can removed
Overview
The Coral M.2 Accelerator is an M.2 module that brings the Edge TPU coprocessor to existing systems and products.
The Edge TPU is a small ASIC designed by Google that provides high performance ML inferencing with low power
requirements: it's capable of performing 4 trillion operations (tera-operations) per second (TOPS), using 0.5 watts for
each TOPS (2 TOPS per watt). For example, it can execute state-of-the-art mobile vision models such as MobileNet v2
at 400 FPS, in a power efficient manner. This on-device processing reduces latency, increases data privacy, and
removes the need for constant high-bandwidth connectivity.
The M.2 Accelerator is a dual-key M.2 card (either A+E or B+M keys), designed to fit any compatible M.2 slot. This form-
factor enables easy integration into ARM and x86 platforms so you can add local ML acceleration to products such as
embedded platforms, mini-PCs, and industrial gateways.
Requirements
The Coral M.2 Accelerator must be connected to a host computer with the following specifications:
For software required on the host, see the software and operation section.
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Speci cations
The design of the M.2 Accelerator adheres to the PCI-SIG's specification for the PCI Express M.2. For in-depth
mechanical details, refer to that specification.
Physical specifications
Host interface
Operating voltage
Environmental reliability
Mechanical reliability
Compliance
requirements.
3 Always handle in static safe environment.
Dimensions
A+E key dimensions
For in-depth mechanical specs, refer to the PCI Express M.2 Specification.
For in-depth mechanical specs, refer to the PCI Express M.2 Specification.
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Figure 2. M.2 B+M key dimensions (in millimeters)
MobileNet v2 0.6 W (7.1 ms @ 141 fps) 0.9 W (3.9 ms @ 256 fps) 1.4 W (2.4 ms @ 416 fps)
Inception v3 0.5 W (58.7 ms @ 17 fps) 0.6 W (51.7 ms @ 19.3 fps) 0.7 W (48.2 ms @ 20.7 fps)
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Thermal limit and operating frequency
The thermal resistance and max allowed temperature of the Edge TPU stack-up is as follows:
The M.2 Accelerator does not include a thermal solution to dissipate heat from the system. In order to sustain
maximum performance from the Edge TPU, it's important that you design your system so the Edge TPU operates well
below the maximum Edge TPU temperature. If the Edge TPU gets too hot, it slowly reduces the operating frequency
and may reset to avoid permanent damage.
The PCIe driver includes a power throttling mechanism (also known as dynamic frequency scaling) and an emergency
shutdown mechanism, based on temperature readings from the Edge TPU. By default, this system checks the Edge TPU
die temperature every 5 seconds and responds as follows:
If the Edge TPU is below 85°C, continue at the "maximum" operating frequency.
If the Edge TPU reaches 85°C, reduce the operating frequency 50% (from "maximum" to "normal").
If the Edge TPU reaches 90°C, reduce the operating frequency another 50% (from "normal" to "low").
If the Edge TPU reaches 95°C, reduce the operating frequency yet another 50% (from "low" to "lowest").
By reducing the operating frequency, the Edge TPU's inferencing speed becomes slower, but it also consumes less
power and hopefully avoids reaching the hardware reset threshold.
As long as the Edge TPU does not reset and the Edge TPU temperature returns to lower levels, the system restores the
operating frequency in the reverse manner—ultimately returning to the maximum operating frequency when the Edge
TPU is below 85°C.
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Connector pinout
GND 75 74 3.3V
NC 73 72 3.3V
NC 71 70 NC
GND 69 68 NC
NC 67 66 NC
NC 65 64 NC
GND 63 62 NC
NC 61 60 NC
NC 59 58 NC
GND 57 56 NC
NC 55 54 NC
GND 51 50 NC
REFCLKn0 49 48 NC
REFCLKp0 47 46 NC
GND 45 44 NC
PETn0 43 42 NC
PETp0 41 40 NC
GND 39 38 NC
PERn0 37 36 NC
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Top side pins Bottom side pins
PERp0 35 34 NC
GND 33 32 NC
NC 23 22 NC
NC 21 20 NC
NC 19 18 GND
NC 17 16 NC
GND 7 6 NC
NC 5 4 3.3V
NC 3 2 3.3V
GND 1
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Figure 3. M.2 A+E key pin positions
GND 75 74 3.3V
GND 73 72 3.3V
GND 71 70 3.3V
NC 69 68 NC
NC 67 66 Key M slot
Key M slot 59 58 NC
GND 57 56 NC
REFCLKp0 55 54 NC
PERp0 49 48 NC
PERn0 47 46 NC
GND 45 44 NC
PETp0 43 42 NC
PETn0 41 40 NC
GND 39 38 NC
NC 37 36 NC
NC 35 34 NC
GND 33 32 NC
NC 31 30 NC
NC 29 28 NC
GND 27 26 NC
NC 25 24 NC
NC 23 22 NC
GND 21 20 NC
NC 11 10 NC
NC 9 8 NC
NC 7 6 NC
NC 5 4 3.3V
GND 3 2 3.3V
GND 1
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.
Figure 4. M.2 B+M key pin positions
The PCIe kernel driver is already upstreamed to kernel.org with source, since version 4.19. For earlier versions, dkms driver
is available via gasket-dkms deb package at https://packages.cloud.google.com/apt coral-edgetpu-stable
main.
To learn how to create models and run inferences the Edge TPU, read TensorFlow models on the Edge TPU.
Version 1.0 (August 2019) Copyright 2019 Google LLC. All rights reserved.