High Density GPUs & FPGAs in Edge Environments

August 23, 2022

High Density GPUs

By Tom Fries, Government/Defense Sales Manager

A June 4, 2020 congressional report, “Intelligence, Surveillance, and Reconnaissance Design for Great Power Competition”, reinforced the importance of the United States’ military continuing to develop ISR (intelligence, surveillance and reconnaissance) capabilities as threats grow around the world. The report states that the goal is to “make rapid sense of that data; securely deliver that data to weapons, weapon systems, and commanders; and possess a workforce that can execute its mission in competition and combat, at a pace greater than the enemy.”  

ISR applications can combine FPGAs and GPUs because they excel when working together in an environment requiring low latency, searching through real time data, providing immediate intelligence to military personnel. ISR equipment often sits in edge transportable tactical command centers near the battlefield. In these scenarios, the designers of the ISR equipment face multiple challenges:  Edge Environments Diagram

  • Optimizing the number of FPGAs and GPUs in limited space
  • Meeting the power requirements of GPUs, and potentially FPGAs with limited power availability
  • Installing high density GPUs and FPGAs behind CPUs with the limitations of server BIOS
  • Accomplishing the above goals with equipment that can meet the stress of an edge transportable environment 

In this post, I will review the unique products One Stop Systems (OSS) offers to meet these challenges. 

Designing the ISR system to allow the GPUs and FPGAs to communicate directly, and thus eliminating the CPU and reducing latency, is important to achieving critical response time requirements. Also, because edge transportable tactical command centers have limited rack space, valuable space can be saved by increasing the density of GPUs and FPGAs behind CPUs -- for example, 8 GPUs and 8 FPGAs behind a single CPU complex. This creates two challenges, however. The first is that most computer BIOS’ are not robust enough to support 16 or more PCIe devices from a CPU. The second challenge is finding a rugged server to support a large number of PCIe devices. 

Added to the above challenges are the mechanical and power requirements of GPUs like the NVIDIA A6000, which are popular in ISR applications. The A6000 is double-wide and requires 300 watts, taxing the thermal and power capabilities of many servers.  

OSS provides practical solutions to meet the challenges of edge transportable environments – limited rack space, robust power and thermal requirements, and rugged systems for harsh environments. Revisiting the requirement of 8 GPUs and 8 FPGAs behind CPUs, OSS offers a rugged short depth server – 3U SDS. This rugged server is only 20” deep, making it ideal for cramped environments. It offers the flexibility of either Intel or AMD CPU Gen4 CPUs, and features two removable canisters that each hold eight 2.5” SATA or NVMe drives. 

3U SDS Server


To address the requirement of putting 8 GPUs and 8 FPGAs behind the CPU, OSS’s 4UP PCIe Gen4 expansion system can be connected to the server via x16 Gen4 links (both copper and fiber cable options). The 3U SDS and 4UP expansion system are a powerful combination.  

The 3U SDS motherboards use a custom OSS BIOS that supports in excess of 120 PCIe devices. So, in our above use case, two 4UPs can be attached to the 3U SDS server, one with 8 GPUs, and the other with 8 FPGAs. This could be taken a step further, with two additional 4UPs providing for 16 GPUs, and 16 FPGAs supported from a single CPU complex. Like the 3U SDS, the 4UP is rugged to handle harsh environments. The 4UP expansion system also has a variant with I/O slots on the front of the chassis, for situations where quick access to PCIe cards from the front of a rack is needed. 

4UP Expansion System

Another strength of the 3U SDS and 4UP expansion system is their versatility in configuring to the requirements of the ISR application. The 3U SDS can stand alone as a powerful compute node with multiple GPUs and/or FPGAs installed in it. Or, multiple 4UP expansion systems can be attached to the 3U SDS. OSS has designed and built compute node configurations with over 20 GPUs and 30 FPGAs connected to one server. 

OSS remains the leader in leading edge PCIe products. Later in 2022, both the 3U SDS and 4UP expansion system will be available as PCIe Gen5.

Sign up for our newsletter, at the bottom of this page, to stay up-to-date with news from OSS!

Click the buttons below to share this blog post!

Return to the main Blog page

Leave a comment

Comments will be approved before showing up.

Also in One Stop Systems Blog

Autonomous Vehicles Industry
Meeting the Data Storage Demands of the Autonomous Vehicles Industry

September 20, 2022

As the forefront of the development of artificial intelligence and a key application within One Stop Systems’ “AI Transportable” industry, autonomous vehicles present an opportunity for innovation in designing ruggedized servers and high-capacity storage products. While none of the vehicles on the road today are truly self-driving, progress is being steadily made towards systems that can successfully predict and navigate everyday traffic. Autonomous cars and trucks being currently deployed are capable of driving with limited operator input, it is predicted that fully autonomous vehicles will be widely available by the end of the decade. One Stop Systems has had the unique opportunity to support this progression by developing products catered to the extensive storage needs of the autonomous vehicles industry.

Continue Reading

OOB System Management
Out-of-Band Management for Compute Systems on the Edge

September 13, 2022

Out-of-Band (OOB) system management is a staple of the modern datacenter. The tools used by administrators have matured over the last 20 years to provide a variety of mission critical features for managing, monitoring, configuring, and updating computer systems independent of the host system’s resources. Intel led the way in the early days by defining a set of interface specifications for basic system management in 1998 called the Intelligent Platform Management Interface (IPMI). Over the next 20 years, 200+ computer vendors joined support for the IPMI specifications. While IPMI has served us well over the years, it is showing its age with a myriad of security concerns and use of legacy standards that make managing modern hybrid infrastructures difficult and complicated. 

Continue Reading

When will automobiles be fully autonomous on the road?
When will Automobiles be Fully Autonomous on the Road?

August 30, 2022

The idea and dream of a fully autonomous car is almost 100 years old. It was first formulated in the US magazine Scientific American. The reason for this was presumably the "American Wonder", a remote-controlled car that drove through Manhattan in 1925. After almost a century, it seems that the automotive industry is on the verge of reaching its goal of having driverless and fully autonomous vehicles participate in everyday traffic.  But when will it finally come? To answer this question, we need to examine the reasons why vehicles are not yet fully autonomous. 

Continue Reading

You are now leaving the OSS website