​Arrcus and UfiSpace Deliver a Production-Ready AI Networking Solution for AMD-Powered AI Infrastructure

Arrcus’s ArcOS® deeply integrated with UfiSpace's Broadcom-based Tomahawk® & Trident open switch platforms creates a high-performance, fully open alternative to proprietary networking fabrics for AMD GPUs & CPUs

SAN JOSE, Calif.--(BUSINESS WIRE)--Arrcus, the leader in distributed networking infrastructure, and UfiSpace, a global leader in open networking solutions, announced the integration of ArcOS® network operating system with UfiSpace's high-performance open switches, powered by Broadcom® silicon. Together, with GIGABYTE’s modular AI Cluster, GIGAPOD featuring the AMD EPYC & AMD Instinct platforms, this solution is showcased at AMD Advancing AI Day 2026 in San Francisco on July 22-23. The solution provides a high-performance, scalable, and pre-validated AI infrastructure designed to meet the demands of large-scale AI training and inference.

An open Ethernet fabric for the AMD AI era

As operators scale AI clusters built on AMD-powered platforms, the network has become the deciding factor in cluster performance, utilization, and cost. Closed, single-vendor fabrics constrain choice and economics while Arrcus and UfiSpace deliver open disaggregated hardware built on cutting edge Broadcom® silicon, with a proven, full-featured NOS – all assembled into a fabric customers can deploy with confidence and operate at scale. This integrated networking solution is the foundation for the high-performance AMD-powered AI rack solution from Giga Computing, a subsidiary of GIGABYTE, giving customers a clear, deployable path from individual switch to full system. This solution was showcased recently at Computex Taipei 2026. (https://www.ufispace.com/company/news/ufispace-and-gigabyte-debut-u-giga-solution-at-computex-2026). GIGABYTE also showcased the solution at the 6th Gen AMD EPYC™ Server CPU platform at AMD Advancing AI Day 2026 in San Francisco on July 22-23.

ArcOS brings scalable multi-tenant lossless-Ethernet (RoCEv2) fabric, congestion-management, advanced load-balancing, and deep-telemetry capabilities AI workloads demand, while UfiSpace delivers the high-density, power-efficient networking hardware — including liquid-cooled, OCP ORv3-ready platforms — to support the most demanding AMD-powered training and inference clusters.

"AI infrastructure needs to be complemented by smart, programmable networking,” said Shekar Ayyar, Chairman and CEO, Arrcus. “By integrating ArcOS deeply with UfiSpace's Tomahawk 6 and Tomahawk 5 Broadcom platforms, we give customers building AMD-powered AI clusters an open, high-performance Ethernet fabric they can deploy with confidence.”

"From the latest Tomahawk® 6 platforms to direct liquid-cooled innovations, UfiSpace’s solutions are engineered to scale with AI," said Vincent Ho, CEO of UfiSpace. "Coupling these platforms with a hardened, L10-validated ArcOS allows customers to accelerate their timeline from evaluation to full production. We are empowering AMD-based environments to unlock their maximum performance, giving organizations the architectural sovereignty to stay agile in a rapidly shifting market."

“We are excited to see Arrcus and UfiSpace bring AI networking solutions powered by Broadcom’s high-density, high-performance networking portfolio,” said Hasan Siraj, vice president of product management, Core Switching Group, Broadcom. “Broadcom is leading the AI networking infrastructure revolution with industry-defining platforms based on our Tomahawk, Jericho, and Trident product lines. We're pleased to see Arrcus and UfiSpace bring this silicon to market in an open, co-validated solution that gives AI infrastructure customers a high-performance Ethernet fabric—from the scale-out compute fabric to storage and management.”

"Deploying AI infrastructure at scale demands incredible engineering agility," said Daniel Hou, General Manager of Giga Computing. "With our GIGAPOD solutions, we have seamlessly paired AMD's latest platform with Arrcus and UfiSpace's open, high-bandwidth switching solutions. This pre-validated solution removes complex integration barriers, allowing our customers to deploy rapidly and lead the generative and agentic AI revolution."

Deep integration for production-ready solutions:

Arrcus and UfiSpace deliver a tightly integrated networking solution that is tested and ready-to-deploy, where switch hardware and network operating system are co-validated, hardened, and optimized together.

• Co-validated and hardened — ArcOS is qualified and performance-tuned on each UfiSpace platform, reducing integration risk and time-to-deployment.
• Optimized for AMD AI workloads — tuned for the traffic patterns, lossless requirements, and scale of AMD Instinct-based clusters.

Built on Broadcom® High-performance AI switch silicon:

At the heart of the fabric is UfiSpace's newest generation of open platforms, powered by the latest Broadcom® switching silicon:

• S9331-64HO and S9331-64HO3L, powered by Broadcom Tomahawk 6 — up to 102.4 Tbps of switching capacity, purpose-built for the radix, bandwidth, and scale of next-generation AI fabrics, including direct liquid-cooled deployments.
• S9321-64EO powered by Broadcom Tomahawk 5 — up to 51.2 Tbps, a proven high-performance fabric and spine platform.
• S9300-32D, powered by Broadcom Trident 4 and S6301-56ST, powered by Trident 3 (TD3) — up to 12.8 Tbps and 130 Gbps respectively, for storage, aggregation, server access, and OOB roles.

About Arrcus

Arrcus is a leading provider of high-performance routing and switching solutions, enabling organizations to achieve superior scalability and reliability. Its ACE platform, powered by the ArcOS network operating system, offers an agile alternative to traditional monolithic systems for datacenter, core, and edge environments.

To meet the distinct demands of AI, Arrcus delivers specialized ethernet fabrics. For training, its scale-out and scale-across solutions handle massive data synchronization and high-bandwidth requirements. For inferencing, the company recently launched the Arrcus Inference Network Fabric (AINF), a policy-driven approach spanning edge, datacenter, and cloud. By reducing token latency and improving throughput, AINF lowers the cost per token and enables sovereign AI, helping global enterprises modernize connectivity for the AI-driven era. The company is headquartered in San Jose, California. For more information, visit the official website or follow Arrcus on LinkedIn and Twitter/X.

About Giga Computing

Giga Computing Technology is an industry innovator and leader in the enterprise computing market. Having spun off from GIGABYTE, we maintain hardware expertise in manufacturing and product design, while operating as a standalone business that can drive more investment into core competencies. We offer a complete GIGABYTE product portfolio that addresses all workloads from the data center to edge including traditional and emerging workloads in HPC and AI to data analytics, 5G/edge, cloud computing, and more. Our longstanding partnerships with key technology leaders ensure that our new products will be the most advanced and they will launch with new partner platforms. Our systems embody performance, security, scalability, and sustainability.

About UfiSpace

UfiSpace is a leading provider of open networking and modular switch platforms, enabling the transformation of telecom, cloud, and data center infrastructure. Backed by deep technical expertise and a global partner ecosystem, UfiSpace delivers scalable, energy-efficient solutions designed to meet the performance and sustainability demands of next-generation networks. As a certified contributor to the Open Compute Project with telecom grade capabilities, UfiSpace plays a key role in accelerating the adoption of open, disaggregated architectures worldwide.
Learn more at www.ufispace.com.
Follow us on our social media channels: LinkedIn, Facebook, and YouTube.