What is a System on Module (SOM)?

A System on Module (SOM) is a compact computing building block that integrates the core elements of an embedded system into a single, ready-to-use module. It typically includes a processor (CPU or SoC), memory (RAM and storage), power management, and essential interfaces. Instead of designing a full board from scratch, engineers build around the SOM to focus on application-specific hardware.

A SOM sits between a System on Chip (SoC) and a single-board computer. It provides a pre-validated computing core that can be integrated into a custom carrier board, making it a modular and scalable approach for embedded system design.

How does a SOM work?

A SOM concentrates all critical computing functions into a standardized module. It connects to a carrier board through high-density connectors that expose key interfaces such as PCIe, Ethernet, USB, serial links, and sometimes high-speed protocols depending on the design.

When powered, the SOM acts as the computing core of the system. It runs the operating system, executes applications, and manages processing tasks. The carrier board provides the application-specific elements: power regulation, connectors, sensors, and any custom hardware needed for the final product.

This architecture allows engineers to upgrade computing performance simply by replacing the SOM, without redesigning the full system. It also reduces development risk since the core module is already tested and validated.

What are the uses of a SOM?

SOMs are widely used in embedded applications where performance, reliability, and long product lifecycles are essential.

In industrial automation, they are used for robotics, control systems, and edge computing platforms requiring real-time processing. In transportation, SOMs power infotainment systems, railway electronics, and advanced driver-assistance systems.

In aerospace and defense, SOMs enable high-performance and modular computing for applications such as signal processing, mission systems, and radar. Their modularity is especially valuable for long-term programs requiring upgradeability.

In medical systems, SOMs are used in imaging devices, diagnostics, and portable monitoring equipment where compactness and reliability are critical.

In AI and edge computing, SOMs support on-device inference for applications like video analytics, smart cameras, and autonomous systems, often leveraging GPU or FPGA acceleration.

Why use a SOM in embedded design?

SOMs reduce development time, simplify hardware design, and improve reliability by separating the computing core from application-specific design. They also enable scalability, allowing product upgrades without a full hardware redesign.

This makes SOM-based architectures a preferred choice for modern embedded systems that require both flexibility and industrial-grade robustness.

That’s why reflex ces offers off-the-shelf modules based on AMD or Altera FPGAs, as well as custom modules tailored to your needs.