Arrcus Powered by NVIDIA BlueField DPUs Enables Scalable and Efficient Networking for Enterprise AI

Arrcus, the hyperscale networking software company and a leader in core-to-edge AI networking solutions, today announced the enablement of AI-scale networking on NVIDIA BlueField-3 DPUs to empower enterprise datacenters with superior scalability, security, and resource efficiency—all while reducing power consumption. By integrating its cutting-edge ArcOS® network operating system with NVIDIA BlueField-3, Arrcus is accelerating software-defined datacenter networking, freeing up critical compute resources, and enabling high-performance workloads.

Also Read: AiThority Interview with Nicole Janssen, Co-Founder and Co-CEO of AltaML

Modern datacenter architectures support multitenancy, allowing a single physical or virtual infrastructure to be securely shared among multiple tenants. This design enables efficient resource sharing, on-demand scalability, and reduced costs. Traditionally, virtualized network functions—such as EVPN/VXLAN, routing, IPsec, Access Control Lists (ACLs), and Network Address Translation (NAT)—are executed on the host CPU to facilitate connectivity. However, Arrcus’ ArcOS enhances efficiency by offloading these critical networking functions to NVIDIA BlueField-3 DPU instead. This shift significantly improves scalability and accelerates performance while freeing up host CPU resources for mission-critical applications. The result is an optimized, high-performance, and cost-effective datacenter infrastructure.

Arrcus ArcOS leverages the NVIDIA DOCA software framework, which provides a broad set of acceleration libraries that enable line-rate, software-defined packet processing and low-latency cloud network connectivity, leveraging purpose-built NVIDIA BlueField hardware accelerators.

Also Read: Why multimodal AI is taking over communication

The combined power of Arrcus ArcOS, NVIDIA BlueField-3 and NVIDIA DOCA delivers:

• Seamless Connectivity – ArcOS accelerated with BlueField and DOCA provides secure overlay tunnels for robust and scalable multi-cloud networking, enabling enterprises to seamlessly extend workloads across environments.
• High-Performance Offloading – Functions like Routing, EVPN/VxLAN, IPSec, ACLs, and NAT are processed on BlueField-3 DPUs, enhancing throughput and efficiency while lowering CPU overhead.
• Optimized Compute Utilization – By reducing the networking burden on CPUs, enterprises can allocate more processing power to other demanding workloads.
• Advanced Traffic Engineering & Security – Built-in QoS, telemetry, and security functions ensure high performance, reliability, and protection against cyber risks.
• Simplified provisioning and automation – Simplified provisioning and agility using NVIDIA DOCA Platform Framework (DPF) to orchestrate, configure and monitor ArcOS on BlueField-3 in cloud-native environments.

“With the exponential rise of AI-driven workloads, enterprises need datacenter networking solutions that deliver high performance, scalability, and energy efficiency without compromise,” said Shekar Ayyar, Chairman and CEO at Arrcus. “By leveraging NVIDIA BlueField-3 with our industry-leading ArcOS, we are transforming software-defined networking and unlocking new levels of efficiency for modern datacenters.”

“As enterprises rapidly adopt AI, the demand for scalable, high performant data center infrastructure has never been greater,” said Ash Bhalgat, Senior Director of AI Networking and Security Solutions, Ecosystem and Marketing at NVIDIA. “ArcOS accelerated with NVIDIA BlueField-3 and DOCA brings new levels of efficiency to enterprise AI factories while supporting zero trust security for agentic AI workloads.”

[To share your insights with us, please write to psen@itechseries.com]