KI plug-in card with Coral Edge TPU for retrofitting.
With the Mustang-T100-T5, ICP Germany expands its portfolio of AI accelerator cards and brings another way to add AI functionality to existing systems.
Powerful, energy-saving and sustainable are three plus points of the new Mustang card. On the MUSTANG-T100, five Coral Edge TPUTM from Google® provide the ability to simultaneously run up to five different AI tasks. The performance of each TPU speaks for itself. 4 trillion computing operations per second at 2 watts power consumption per TPU, 2 TOPS per watt, in total according to Adam Riese up to 20 TOPs at 15 watts power consumption of the card. Instead of replacing existing systems with new ones, they can be sustainably upgraded with the Mustang-T100-T5.
The five Coral Edge TPUs fit on a half-height, half-length low-profile PCI Express card. In existing systems with PCI Express x4 slot, the Mustang-T100 can thus be easily integrated. In addition to x86 platforms, ARM platforms are also supported, as well as Windows operating systems and Linux. Upgrading existing systems thus enables a longer runtime and a reduction of the CO² footprint..
Easy implementation of an AI model is guaranteed by Tensorflow Light. Systems with Mustang TPU henceforth offer parallel multitasking capability with which a separate AI model can be computed per TPU. Shorter latency times can be achieved by model pipelining. This splits the AI model into segments to run the computation in parallel on each TPU, reducing the overall computation time. AutoML Vision Edge accelerates transfer learning, the execution of pre-trained models directly on the system, by using the TPUs. By transfer learning and optimizing the system with a small data set, accuracy is increased and training time is significantly reduced. Image classification, object recognition, and image segmentation are ideal applications for the Mustang-T100.
- ● Retrofittable AI accelerator card
- ● Low-profile PCI Express form factor
- ● 5 Google Coral Edge TPUsTM
- ● 2nd generation PCI-Express x4
- ● Multiple card support
- ● 15 Watt Stromverbrauch
- ● Inference Systeme
- ● AIoT Edge Computer
- ● Machine Learning Applikationen
- ● Object recognition
- ● Face recognition