IEI Mustang-T100-T5: Features
- 5 x Google Edge TPU ML accelerator
- 20 TOPS peak performance (int8)
- Host interface PCIe Gen2 x 4
- Low-profile PCIe form factor
- Supporting use of multiple cards
- 15W TDP
- RoHS compliant
The Next Wave for Edge AI Computing with Coral Edge TPU
In recent years, artificial intelligence (AI) has been applied in various industries and has consequently opened up further opportunities and provided efficiencies in lots of different applications. Edge AI has become a virtual role with customers spending more time in deploying services to local devices in order to increase privacy, security and minimise latencies.
IEI’s Mustang-T100 accelerator card leverages the powerful Coral technology, a complete and fast toolkit for building products with local AI, to allow AI developers to access different AI inference platforms at the Edge.
The Mustang-T100 integrates five Coral Edge TPU co-processors in a half-height, half-length PCIe card and offers computing speeds up to 20 TOPS and low power consumption (only 15W). Powered by a well-developed Tensorflow Lite community, the Mustang-T100 is an ideal AI PCIe accelerator card for image classification, object detection and image segmentation.
High Compatibility from The Start
To support the diverse needs of IT/AI developers, the Mustang-T100 can be implemented for various operating systems, such as Linux and Windows, and can be deployed for both X86 and ARM hardware platforms to accelerate and maximum Edge AI performance. Additionally, and combined with TensorFlow Lite, there is no need to build ML training models from the ground up. TensorFlow Lite models can be compiled to run on the Edge TPU, completely.
Multitasking or Pipelining – Select Your Inferencing Mode
For numerous AI applications at the Edge, clients can select from two different modes to run their inferencing project depending on their needs.
Multitasking or Pipelining – Select Your Inferencing Mode
If there’s a need to run multiple models, each model can be assigned to a specific Edge TPU and ran in parallel at the same time for extreme computing performance.
Model pipelining to get faster throughput and low latency
For other scenarios that require very fast throughput or large models, pipelining your model allows you to execute different segments of the same model on different Edge TPUs. This can improve throughput for high-speed applications and can reduce total latency for large models.
Optimize Pre-trained Models at The Edge
The Mustang-T100 supports AutoML Vision Edge that AI developers can use Edge TPU to accelerate transfer-learning with a pre-trained model at Edge devices, directly. The models can be optimised to increase model accuracy and save training time by running fewer than 200 images.
Easily Create an AI Environment
IEI provides AI-ready IPC hardware platforms, such as the TANK series, FLEX series and the DPRC series. All these series are ideal for deep learning inference computing to help get faster, deeper application insights. All IEI AI-ready systems are able to support up to four Mustang-T100’s to deploy and run your AI applications efficiently.
Application Field Installs
Reliable University Campus
Security
Because university campus’ can be large, therefore requiring numerous cameras to ensure the entire campus is covered for multiple scenarios concerning the safety of the campus – and those people that work on & attend it, there needs to be a high level of computing performance present.
Using the DRPC IPC series configured with the Mustang-T100 solution to cope with the required AI workload, multiple algorithms and multiple camera channels can be supported.
University Campus’ are using this system for face recognition, unauthorised personnel alerts, crowd management, people count and parking management.
Intelligent Traffic
Management
For pedestrian safety and traffic efficiency, AI can help Traffic Control Centres to monitor road-flow, pedestrian crossings and bus stops to provide real-time reaction to live traffic conditions. The system is accelerated by the Mustang-T100, powered by Coral Edge TPU. With 5 cores of TPU onboard, the system affords itself multiple camera channels for observation.
SMART Retail
Management
Various in-store retail monitoring applications increase on a year-by-year basis. However, with the recent COVID-19 period across the globe, new requirements concerning the gauging of customer footfall in compliance of social distancing, correct fitting of face and restriction of instore numbers of customers – has become a new requirement.
By configuring the FLEX series with the powerful Mustang-T100 accelerator card, more camera channels have been enabled for deployment across both small, local supermarket branches – and larger superstores – to track, monitor and enforce safety measures for retail customers at any given time.
Specification
| Form Factor | |
|---|---|
| Form Factor | Low-profile PCIe card |
| System | |
| CPU | 5 x Google Coral Edge TPU |
| Cooling method / System Fan | Active Fan |
| Physical Characteristics | |
| Dimensions (LxWxH) (mm) | 167.64 (L) x 64.41 (W) x 18 (H) |
| Environment | |
| Operating Temperature | -20°C ~ 55°C |
| Humidity | 5% ~ 90% RH, Non-condensing |
| Certifications | |
| Safety & EMC | RoHS |
We like to make life easier ….
BVM supply a wide and diverse range of Industrial and Embedded Systems.
From Industrial Motherboards, SBCs and Box PCs, to Rack Mount computers and Industrial Panel PCs.
Our technical & commercial team members will always provide you with valuable but impartial advice around the products and services that BVM provide. With their collective backgrounds, they’ll provide you with the benefit of their knowledge and experiences when & where you need it. We’ll always help you in the first instance and get back to you when additional information is required.
You can either call us directly on +(0) 1489 780 144 and talk to one of the team | E-mail us at sales@bvmltd.co.uk | Use the contact form on our website