Join Premio at Booth #1477C during the 2019 NAMA Show, Apr 24 – 26 in Las Vegas Convention Center where automatic merchandising professionals meet and explore the latest innovations to improve business efficiency for self-service machines and micro markets.
Industrial GPU Computer
An Industrial GPU computer is a rugged, edge computing solution that utilizes a dedicated GPU (Graphics Processing Unit) card for computational intensive applications requiring advanced image processing and machine vision/intelligence. In addition to a performance boost from the GPU, an industrial GPU computer will also have all the key requirements of what makes up an industrial computer, such as support for high shock/vibration ratings, wide operating temperatures, and over voltage/reverse polarity protection. The industrial GPU computer is transforming how factory automation and smart manufacturing verticals are utilizing computational analysis and machine intelligence in real-time.
Computer / Machine Vision
- Production line inspection
- Medical imaging
- Metrology imaging
- Facial recognition
- Autonomous vehicle
- Factory automation
- Deep learning
- Machine learning
Both GPUs (Graphics Processing Unit) and CPUs (Central Processing Unit) have many traits in common; they are both specifically designed microprocessors to handle various tasks. The key difference is HOW they process these tasks. In a computer, the CPU is often referred to as the brain of the system; it is the central processing unit which handles all computing tasks. A GPU, while similar to a CPU, is engineered specifically to process or render graphics. As such, a GPU can work in conjunction with the CPU to help it offload graphics intensive tasks, while freeing up the CPU for other non-graphics related jobs.
The CPU is ideal for serialized, generic tasks, which makes it well-suited for common business or productive applications such as Word, Excel, or a web browser. The number of cores in a given CPU is limited, up to 28 cores for the latest Xeon server class CPU. Conversely, the GPU comes with hundreds or even thousands of cores, which is designed for parallelized, specific tasks; the GPU is optimized for intensive computational applications such as image processing or AI.
As an example, let's examine the fundamentals of image processing. The 4K image of the clock consists of 8.2 million discrete pixels (4K resolution is 3,840 x 2,160, which gives us 8.2 million pixels). From a high level, in order to process the image, we will need to perform some type of computation to each individual pixel. If this task is given to a CPU, with its limited number of cores, the processing time will take very long, as the CPU does not have enough cores to handle the task in parallel. The GPU, which its thousands of specialized cores, can complete the task up to 50-100X faster due to its parallel architecture. This makes the GPU the optimal microprocessor to handle tasks that require parallelism with a high degree of computation.
We begin with a proven embedded system that is engineered to withstand extreme shock and vibration, along with a wide operating temperature range; the inclusion of an industrial class GPU will enable the system to operate reliability in industrial/manufacturing sectors, with the GPU handling AI or image processing applications that require massive parallelism.
One way to compare the performance of various GPUs is by their TFLOP rating. TFLOP stands for "teraflop," which is a measurement of the GPU to performance one trillion floating point operations per second. By adding a GPU to an industrial system, another critical variable we need to consider is the TDP (Thermal Design Power) rating of the GPU. The TDP rating tells us the maximum heat, in watts, generated by the GPU when operating at maximum capacity. This is key in determining the operational temperature range of an industrial GPU system, which should be optimally in the range of -25C to 60C for factory/manufacturing averse conditions. By looking at the Performance versus TDP chart below, we can see the direct correlation between these two values: as the GPU performance increases, so does its corresponding TDP value. When designing an industrial GPU system, there is a fine balance to strike between GPU performance versus overall system operating temperature.
|Ports Accessibility||All front facing ports
Ideal for factory line mounting
|Front and rear facing ports|
|Vibration Rating||3 Grms
Ideal for staionary factory line mount
|Higher 5 Grms
Ideal for mobile mount in cars or carts
|Wide Range Temperature||-25C to 70C (-13F to 158F)||-25C to 70C (-13F to 158F)|
|Wide Range Voltage||9 ~ 50 VDC||9 ~ 48 VDS|
|Expandability||2x mini PCIe slots||2x universal slots for additional ports, 2x mini PCIe slots|
|Communication||Wifi optional; 2x antenna holes||Wifi and Cellular (4G/LTE) optional; 4x antenna holes|
|Mounting||Wallmount, Bookmount, or DIN rail||Wallmount|
|Processor||Intel Skylake 6th gen / Kaby Lake 7th gen i3, i5, i7, quad core|
|Digital I/O ports||8 in / 8 out isolated|
|Video Ports||2x DVI-I, 1x HDMI, 3x DisplayPort|
High Performance Edge Computing
By moving computing and data closer to the user, edge computing can provide significantly improved throughput, better performance, and real-time processing. The ability to analyze data closer to the source will minimize latency, reduce network traffic, and lower data management costs while improving privacy and security.
Read More on the benefits of Edge Computing ...
Reduce network traffic
and cloud reliance
Accelerate time to action
Built on NVIDIA® Pascal™ GPU architecture
Premio GPU Computing Systems are built on the highly efficient NVIDIA® Pascal™ architecture featuring 4GB GDDR5 and 768 CUDA® cores that delivers amazing performance with impressive power-efficiency. The GTX 1050Ti features 1.73 TFLOPS with a max TDP of 75W, making it the best fit for industrial computers that require optimal performance and wide range operating temperature.
|NVIDIA® CUDA® Cores||768|
|Frame Buffer||4 GB GDDR5|
|Memory Speed||7 Gbps|
|Boost Clock||1392 MHz|
Support up to 8K Resolution & 6 Independent Displays
Through combining discrete NVIDIA GTX 1050Ti and integrated Intel HD Graphics units, Premio GPU computers enable support for up to 8K resolution and six independent displays.
Versatile Industrial I/O Interfaces & Diverse Expandability
Rich multi-functional I/O ports including Dual GbE (additional 4x LAN port on RCO-6020-1050TI), 4x COM, up to 8x USB, and 8 digital I/O ports allow Premio’s industrial GPU computer for better expansion and future utility. An additional up to 4 mini PCIe slots provide more flexibility for multiple applications.
Robust system design enabling GPU computing in harsh environment
Flexibility and performance are packed into a robust and durable design for incomparable user experience. Premio GPU computing systems go through a rigorous design cycle with strict quality assurance procedures and extensive product testing. These industrial computers are the ideal solution for versatile operations in rough and harsh environments.
Successful Deployment Applications For Industrial GPU Computer
Reducing Airport Security Check to Just 20 Seconds using powerful GPU computing for machine learning and facial recognition
Performance GPU Computer Helps Industrial Metrology Increase Scan Speeds up to 10X Faster