アナログ・テック List of Products and Services

The AT-IPCFG series from Analog Tech deploys image AI and edge AI for harsh environments where computers struggle, such as safety management at construction, agriculture, and surveying sites, inspection sites for roads, tunnels, and bridges, vehicles like unmanned transport vehicles, and environments with high temperatures, low temperatures, high humidity, and dust. Equipped with a ruggedized housing that enhances environmental performance, it includes AI chips and AI modules for inference. It provides low-latency, high-speed image AI processing on edge devices, including high-resolution image processing, video processing from multiple cameras, real-time image processing, and AI inference processing. Custom solutions and OEM supply tailored to customer applications are available, so please feel free to consult with us.

The AT-IPCWG series is a compact wall-mounted industrial PC that can accommodate a variety of NVIDIA GPUs ranging from mid to high-end. Customization and OEM options are available to suit customer needs, so please consult with our consultants. 【Features】 ■ Equipped with NVIDIA Geforce/T1000/T4/Quadro Thanks to excellent cooling performance, it can support various NVIDIA GPUs. It is compatible with the relatively affordable Geforce series for gaming, the T4 for data centers, and the Quadro series for image processing and simulation. You can also choose fanless GPUs like the T4 or passive thermal types of Quadro. ■ Expandable with various functional boards including image input boards A chassis designed for enhanced expandability is available. It supports various input and output options tailored to your needs, such as expanding inputs for Camera Link and CoaXpress. ■ Installable on-site with dust filter Standard equipped with a dust filter. It boasts excellent environmental resistance and can be deployed in locations where rack mounting is difficult. 【Further features continue in the section "Basic Information" below】

The AT-GPU series features AMD EPYC 7002 series processors and supports NVIDIA's latest Ampere architecture GPUs with PCI Express 4.0. It is a cost-effective limited production model of a supercomputer designed for AI. 【Features】 ■ PCI Express Gen 4 Support Equipped with AMD EPYC 7002 series processors, it supports six "PCI Express Gen 4 (x16)" slots and two "PCI Express Gen 4 (M.2)" slots. It supports various high-speed communications, including the latest NVIDIA GPUs, RAID, SSDs, and high-speed NICs. 【More features continue in the "Basic Information" section below】

Our company manufactures industrial PCs and servers tailored to customer requests, and we have prepared base models that embody various customer demands and the latest technologies in advance. If the specifications match, you can use them as they are. If the specifications do not match, we can customize the parts of the base model that do not fit to shorten the specification design process and reduce the lead time for delivery to customers. We will start by confirming base chassis that are close to the customer's required specifications. If you have a chassis that resembles your required specifications, please feel free to consult with us. For the main lineup of base models, please refer to the basic information below.

■What is an AI Chip■ An AI chip is a semiconductor designed to accelerate AI computational processing. In image processing and AI tasks, GPUs are becoming almost standardized for training purposes. However, the specifications differ between training and inference, and inference can be achieved using not only GPUs but also FPGAs and ASICs. In particular, inference for edge applications is implemented in devices that process AI on-site alongside IoT, necessitating considerations for adoption based on not only "computational performance" but also "low power consumption," "wide operating temperature," and "long supply duration" for embedded implementations in devices. ■Introduction of AI Chips Suitable for Industrial Applications■ Hailo's AI chip boasting low power consumption and high inference processing capability: 'Hailo-8 AI Processor' The Hailo-8 processor is a chip specifically developed by Hailo for AI processing. Among AI chips, it has the following features that make it particularly suitable for industrial applications: (Feature 1) AI processing with high power efficiency! Power efficiency of 3 TOPS/W (Feature 2) High environmental durability! Operating temperature range of -40°C to 85°C (Feature 3) Planned for long-term supply! Expected supply duration of up to 10 years for the chip alone.

■Challenges of Implementing Image AI and Edge AI■ Recently, the development of various image AI and edge AI applications has progressed, moving into the device implementation phase. In this phase, unlike in offices or data centers, devices are required to have stringent environmental durability, which presents a challenge for the implementation of edge AI. On the other hand, the data processed at the edge continues to increase in volume and real-time demands, leading to a linear rise in component performance, resulting in extreme heat generation from CPUs, GPUs, and AI chips, which adds to the challenges of device implementation. In edge AI, to enhance processing performance, it is crucial to address heat dissipation from high-heat-generating components such as GPUs and AI chips while ensuring extremely strict environmental durability. This involves temperature control, vibration mitigation, and mechanical design, which are important factors for implementing image AI and edge AI applications in various settings. ■High Environmental Durability Edge AI Platform■ Analog Devices provides a platform for implementing AI in harsh field environments such as 'high temperature/low temperature, water, high humidity, and dust' by equipping industrial PCs with AI chips that excel in environmental durability.

1. Equipped with Hailo's AI chip 'Hailo-8 AI Processor' Hailo is a chip manufacturer established in 2017 that develops high-performance AI processors for edge devices. The AI chip 'Hailo-8' developed by the company places data pools and computing units in a very small area, achieving low latency and low power consumption processing without using external interfaces or I/O. It reduces the area to one-fifteenth compared to general-purpose GPUs, achieving 20 times the power efficiency, and achieves 1.1 times the performance in FPS comparisons. ● High performance of 26 TOPS per chip (when used as a standalone chip or M.2 module) with top-class AI processing performance at a power efficiency of 3 TOPS/W ● Executes AI inference at the edge in real-time, with low latency and high efficiency ● With Hailo's data flow compiler and support for standard AI frameworks, developers can easily and quickly port neural network models.

1. Equipped with NVIDIA's Volta AI Architecture system-on-module NVIDIA Jetson Xavier NX, featuring 384 CUDA cores and 48 Tensor cores, capable of executing multiple state-of-the-art neural networks in parallel at up to 21 TOPS. Existing NVIDIA customers can implement deployments that directly inherit their existing training data and development assets. 2. 8-Port PoE (Power Over Ethernet) for Image AI Equipped with a total of 75W 8-port PoE, allowing direct power supply to cameras connected via Ethernet ports. 3. Operating support from -25°C to 55°C Supports operation in a wide temperature range from -25°C to 55°C. 4. Base model customizable for OEM If specifications do not match, we can customize the non-conforming parts according to the situation, aiming to shorten specification design and lead time for customer delivery. We will begin discussions by confirming the base chassis that closely aligns with customer requirements.

1. Hailo's AI chip Hailo-8 Deep Learning Processor equipped Hailo is a chip manufacturer established in 2017 by members of the elite technology unit of the Israel Defense Forces, developing high-performance AI processors for edge devices. The Hailo-8 is a chip designed with an architecture specialized for AI processing developed by Hailo. As a result, it can place data pools and computing units in a very small area, enabling processing with low latency and low power consumption without using external interfaces or I/O. Compared to general-purpose GPUs, it reduces the area to one-fifteenth and achieves 20 times the power efficiency, attaining 1.1 times the performance in FPS comparisons. ● High performance of 13 TOPS per chip (when using Mini-PCI module) with top-class AI processing performance at 3 TOPS/W power efficiency ● Executes AI inference at the edge in real-time, with low latency and high efficiency ● With Hailo's data flow compiler and support for standard AI frameworks, developers can easily and quickly port neural network models.

1. Hailo's AI chip Hailo-8 Deep Learning Processor equipped Hailo is a chip manufacturer that develops high-performance AI processors for edge devices, founded in 2017 by members of the elite technology unit of the Israel Defense Forces. The Hailo-8 is a chip with an architecture specialized for AI processing developed by Hailo. As a result, it can place data pools and computing units in a very small area, enabling processing with low latency and low power consumption without using external interfaces or I/O. Compared to general-purpose GPUs, it reduces the area by a factor of 15 and achieves 20 times the power efficiency, achieving 1.1 times the performance in FPS comparisons. ● High performance of 13 TOPS per chip (when using Mini-PCI module) with top-class AI processing performance at 3 TOPS/W power efficiency ● Executes AI inference at the edge in real-time, with low latency and high efficiency ● With Hailo's data flow compiler and support for standard AI frameworks, developers can easily and quickly port neural network models.

This is a compact industrial computer equipped with an NVIDIA T1000 GPU card, an industrial PC. It has a drive bay for 2.5" HDD/SSD (compatible with industrial SSDs). It is equipped with isolated DIO, and a DIN rail can be installed as an option, allowing for combinations with industrial power supplies.

This is a compact industrial computer equipped with an NVIDIA RTX A2000 GPU card, an industrial PC. It has a 2.5" drive bay for HDD/SSD (compatible with industrial SSDs). It is equipped with an isolated DIO interface and can accommodate an additional PCIe (x1) card. A DIN rail can be optionally installed, allowing for combination with industrial power supplies.

1. Equipped with NVIDIA Tesla T4, a high-end GPU from NVIDIA The "NVIDIA Tesla T4" is utilized in various fields that require high processing power, such as hyperscale data centers, deep learning training, inference, machine learning, data analysis, and graphics. 2. Industrial GPU computer with dust filter, suitable for production environments Despite being wall-mounted, it achieves compactness through excellent cooling design. It comes standard with a dust filter, and optional industrial-grade parts like industrial SSDs can be configured. It boasts superior environmental resistance, making it suitable for environments where rack mounting is difficult. It is also ideal for image processing and AI inference in applications requiring environmental durability, strongly promoting edge computing in the field. 3. Customization and OEM available It offers optional general interfaces such as RS232/422/485 and DIO, and is a model that can provide configuration proposals based on specific requirements and OEM supply. Please consult our consultants regarding integration into the products you manufacture. [Continues with "Basic Information" below]

1. Supports both passive and active thermal solutions for all NVIDIA Tesla/Quadro/A Series GPU cards. With excellent cooling performance, it can accommodate various NVIDIA GPUs. This device can support high-end GPUs for data centers, such as the Quadro RTX 8000 and 6000 series, as well as the A series, using passive thermal solutions. Additionally, this device is suitable for on-premises edge applications within edge computing that interacts with the cloud, allowing for processing of highly confidential data analysis at the edge, not just inference but also data analysis that may raise concerns in the cloud. 2. High expandability allows for the addition of various functional boards, starting with image input boards. With high expandability, it is possible to enhance performance for image processing by adding boards such as Camera Link and CoaXpress. It is suitable for real-time image processing applications using machine vision. [Continued in the "Basic Information" section below]

This is a wall-mounted computer suitable for image AI and edge AI, capable of accommodating NVIDIA RTX 4000 Ada or NVIDIA RTX 4500 Ada GPU cards. In addition to long-term supply support, it features an interface for external switches, the adoption of SSDs with Power Loss Protection (PLP) functionality, an isolated DIO interface, and supports DC input from 9 to 48V, providing the robustness and expandability required for industrial applications.

This is a compact industrial computer equipped with NVIDIA RTX A5000/RTX A4500/RTX A4000 GPU cards. It complies with medical electrical equipment standards EN 60601-1-2:2015, EN60601-1:2014, and FCC: Part 18 Class B (excluding add-on cards), offers long-term supply capability, is equipped with protective installation terminals, and optionally provides capture cards (SDI/HDMI/DVI).

It is a compact fanless industrial computer. It offers long-term supply capability, features an external switch interface for embedded applications, operates within a temperature range of -20°C to 50°C, uses an SSD with PLP functionality, supports DC input of 9 to 30V, and is SIM compatible (optional).

It is a compact fanless industrial computer. It offers long-term supply capability, equipped with an external switch interface for embedded applications, operating temperature range of -20°C to 50°C, uses SSD with PLP functionality, features isolated DIO interface, supports DC input of 9 to 30V, and is SIM compatible (optional).

It is a compact fanless industrial computer. Long-term supply is possible, equipped with an external switch interface for embedded applications, operating temperature range -40℃ to 65℃, uses SSD with PLP function, equipped with isolated DIO interface, supports DC input of 9 to 48V, SIM compatible (optional).

It is a compact fanless industrial computer. Long-term supply is possible, equipped with an external switch interface for embedded applications, operating temperature range -40°C to 65°C, uses SSD with PLP function, equipped with isolated DIO interface, supports DC input of 9 to 48V, SIM compatible (optional).

This is a compact fanless industrial computer. Recommended industrial PC for Edgecross, capable of long-term supply, equipped with an external switch interface for embedded applications, features 2 PoE ports, includes a 2.5" SSD swap tray (with lock), operating temperature range of -40℃ to 65℃, adopts SSD with PLP functionality, equipped with isolated DIO interface, supports DC input of 9 to 48V, and SIM compatible (optional).

It is a compact fanless industrial computer. Long-term supply is possible, supports external switch interfaces for embedded applications, operating temperature range -25°C to 55°C, equipped with SSD featuring PLP functionality, includes isolated DIO interface, supports DC input of 9 to 48V, and SIM support (optional).

It is a compact fanless industrial computer. It supports long-term supply, has an interface for external switches suitable for embedded applications, is equipped with four PoE (IEEE802.3at) ports, operates in a temperature range of -20℃ to 65℃, uses SSDs with PLP functionality, features a 2.5-inch hot-swap bay, includes an isolated DIO interface, and supports DC input of 9 to 48V.

The AT-GPU series features AMD EPYC 7002 series processors and supports NVIDIA's latest Ampere architecture GPUs with PCI Express 4.0. It is a cost-effective limited production model supercomputer designed for AI applications. ■ PCI Express Gen 4 Support Equipped with AMD EPYC 7002 series processors, it supports six "PCI Express Gen 4 (x16)" and two "PCI Express Gen 4 (M.2)" connections. It supports various high-speed communications, including the latest NVIDIA GPUs, RAID, SSDs, and high-speed NICs. ■ NVIDIA® Ampere Support It supports high-end GPUs for workstations and data centers, such as the "NVIDIA RTX A6000" and "A100-40G," as well as budget-friendly (gaming-oriented) GPUs like the "NVIDIA Geforce RTX 3080" and "NVIDIA Geforce RTX 3090." It accelerates various AI and HPC applications, reducing the time required for AI training and inference.

This is a wall-mounted industrial computer. It offers long-term supply capability, operates within a temperature range of -25°C to 65°C, is equipped with a 2.5-inch hot-swap bay, features an isolated DIO interface, supports DC input from 9 to 50V, and has a DIN rail bracket option.

It is a highly versatile rack-mounted industrial computer. Long-term supply is possible, equipped with one PCI Express (x16) slot, supports triple monitor output, and offers a rich variety focused on versatility (2.5" drive swap-bay, mirroring unit).

This is a 2U rack-mounted industrial computer equipped with NVIDIA GPU cards. It offers long-term supply capability, supports the installation of three NVIDIA RTX A2000 cards, PCI-Express Gen4 support, and Intel Optane Persistent Memory support (optional).

Standard 4U rack-mounted industrial computer, industrial PC. A low-cost standard model available for long-term supply, ivory color for factory automation, with a wide variety of options focusing on versatility (2.5" drive swap-bay, mirroring unit), and capable of accommodating full-length/full-height add-on cards. Secures expansion drive space with three 5.25-inch bays.

High-reliability/high-performance 4U rack-mounted industrial computer, industrial PC. Long-term supply available, depth of 408mm, capable of accommodating full-length/full-height add-on cards, with a wide variety of options focused on versatility (2.5" drive swap-bay, mirroring unit).

This is a 4U rack-mounted industrial computer equipped with an NVIDIA GPU card. It offers long-term supply, has a depth of 408mm, supports RTX A4000/A5000/A6000 (up to 3 cards can be installed), supports PCI-Express Gen4, and supports Intel Optane Persistent Memory (optional).

This is a 4U rack-mounted industrial computer equipped with an NVIDIA GPU card. It offers long-term supply capability, a depth of 550mm, supports NVIDIA A16/A30/A40/A100 (up to 3 cards can be installed), PCI-Express Gen4 support, and Intel Optane Persistent Memory support (optional).

This is an industrial PC equipped with NVIDIA RTX A4000 and an integrated touch panel. It offers long-term supply, features a 27-inch TFT LCD, complies with IP54 on the panel side, and an optional capture card is also available.

We have packaged a development environment to easily start developing AI applications using the Hailo-8 AI Processor. The evaluation development kit includes PC hardware equipped with a Hailo-8 Mini-PCIe module (with a performance value of 13 TOPS) as the execution environment for running Hailo-8. ■ Deployment toolchain compatible with industry-standard frameworks It supports model conversion from industry-standard frameworks to Hailo format, converting the model's bit configuration to Hailo's configuration using the latest quantization algorithms. It allocates hardware resources and user network resources to the physical resources of the Hailo device. The model is compiled into a Hailo binary, the binary is loaded, and inference is executed on the Hailo target device. The SDK supports both standalone inference, which allows direct access to the device, and TensorFlow integrated inference, which facilitates integration with existing environments.

1. IP65 Waterproof Performance Equipped with IP65-rated dust and waterproof performance, and featuring an excellent heat dissipation design with a fanless chassis, it can be implemented in various environments, including outdoor settings. It supports heavy-duty environments with various interfaces equipped with industrial waterproof connectors compatible with M12. It is possible to replace parts with industrial-grade components such as industrial SSDs. 2. Base Model Capable of Customization and OEM If the specifications do not match, we can customize the incompatible parts according to the situation, aiming to shorten the specification design process and reduce the lead time for customer delivery. We will start by examining base chassis options that are closest to the customer's required specifications.

1. Diverse expansion capabilities. 【Full-length / Full-height add-on cards can be installed】 Full-length / full-height add-on cards can be installed in a compact chassis, allowing for various I/O and function expansion boards tailored to customer applications. 【Optional modules】 Various optional modules such as a 2.5" Drive Swap-bay and mirroring units can be added. 【Standard equipment】 Equipped with GPIO as standard. 2. Installable on-site with a dust filter. The AT-IPCW001 can be deployed in harsh environments such as production sites, strongly promoting edge computing in various production settings. It excels in environmental performance and can be introduced in locations where rack mounting is difficult. It strongly promotes edge computing in applications requiring environmental performance, such as medical, retail, and transportation.

It is a wall-mounted industrial computer. Long-term supply is possible, operating temperature range -25°C to 65°C, equipped with a 2.5-inch hot-swappable bay, features an isolated DIO interface, and supports DC input of 9 to 50V.

1. Equipped with two Hailo AI chips, supporting multi-AI algorithms The AT-IPCFG004 can apply multiple algorithms simultaneously to the source by incorporating several Hailo-8 AI chips. By applying different types of learning models, such as area recognition and object recognition, or object recognition and pose estimation, to the input source at the same time, it expands the AI inference capabilities. 2. Equipped with Hailo's Hailo-8 AI processor Hailo is a chip manufacturer established in 2017 that develops high-performance AI processors for edge devices. The Hailo-8 AI processor developed by the company places the data pool and computing units in a very small area, achieving low latency and low power consumption processing without using external interfaces or I/O. ● With a top-class AI processing performance of 26 TOPS per chip (when used as a standalone chip or with an M.2 module) at a power efficiency of 3 TOPS/W ● Executes AI inference in real-time, with low latency and high efficiency ● Hailo's SDK allows for easy porting of neural network models

■Equipped with two 'Hailo-8 AI processors' in a single fanless chassis■ The 'AT-IPCFG004' is an industrial computer with excellent environmental performance, featuring two high-performance AI chips designed for industrial applications. It can simultaneously apply different learning models, such as area recognition and object recognition, as well as object recognition and pose estimation, to input sources, thereby expanding AI inference capabilities. ■Equipped with two high-performance AI chips from Hailo, supporting multi-AI algorithms■ In AI deep learning, depending on the 'types of data sources' being handled, such as images, audio, and time series, as well as the 'analysis purposes' for what you want to learn from that data source, suitable algorithms are selected and applied accordingly for tasks like classification, depth estimation, object recognition, face detection, and pose estimation. Generally, in edge inference, one learning model is executed for one image source. With the AT-IPCFG004, by incorporating multiple Hailo-8 processors, it has become possible to 'simultaneously apply multiple algorithms to the source.'

1. Equipped with Hailo's AI chip 'Hailo-8 AI Processor' Hailo is a chip manufacturer established in 2017 that develops high-performance AI processors for edge devices. The AI chip 'Hailo-8' developed by the company places data pools and computing units in a very small area, achieving low latency and low power consumption processing without using external I/F or I/O. It reduces the area to one-fifteenth compared to general-purpose GPUs, achieving 20 times the power efficiency, and reaches 1.1 times the performance in FPS comparisons. ● High performance of 26 TOPS per chip (standalone chip, when using M.2 module) with top-class AI processing performance at a power efficiency of 3 TOPS/W ● Executes AI inference at the edge in real-time, with low latency and high efficiency ● Easily and quickly ports neural network models with Hailo's data flow compiler and support for standard AI frameworks 2. Equipped with four PoE ports Supports IEEE802.3at, providing up to 25.5W across four ports and up to 30W on a single port. It allows direct power supply to IP cameras from a PC.

1. Equipped with two Hailo AI chips, supporting multi-AI algorithms The AT-IPCFG004 can apply multiple algorithms simultaneously to a source by incorporating several Hailo-8 AI chips. By applying different types of learning models, such as area recognition and object recognition, or object recognition and pose estimation, to the input source at the same time, it expands the AI inference capabilities. 2. Equipped with Hailo's Hailo-8 AI processor Hailo is a chip manufacturer that develops high-performance AI processors for edge devices, established in 2017. The Hailo-8 AI processor developed by the company places data pools and computing units in a very small area, achieving low-latency and low-power processing without using external interfaces or I/O. ● High-class AI processing performance at 3TOPS/W, achieving 26TOPS of high performance per chip (when used as a standalone chip or with an M.2 module) ● Executes AI inference in real-time, with low latency and high efficiency ● Easily ports neural network models using Hailo's SDK

1. Equipped with two Hailo AI chips, supporting multi-AI algorithms -------------------------------------------------------------------- By incorporating multiple AI chips, it is possible to apply multiple AI algorithms simultaneously to the input source. This expands the AI inference capabilities by applying different types of learning models, such as "area recognition and object recognition" or "object recognition and pose estimation," to the input source at the same time. 2. IP65 waterproof performance -------------------------------------------------------------------- Adopting dust and waterproof performance compliant with IP65, along with a fanless design featuring excellent heat dissipation, it can be deployed in various environments, including outdoors. Various interfaces equipped with industrial waterproof connectors compatible with M12 ensure adaptability to harsh environments. Additionally, it is possible to replace components with industrial-grade parts such as industrial SSDs.