Rocky Linux from CIQ adds NVIDIA DOCA OFED for rapid GPU deployment in data centers

CIQ has announced that Rocky Linux from CIQ (RLC) and Rocky Linux from CIQ – AI (RLC-AI) now provide integrated support for the NVIDIA DOCA OpenFabrics Enterprise Distribution (OFED) stack, supplementing its existing out-of-the-box NVIDIA CUDA Toolkit integration. CIQ reports that RLC is the first Linux distribution licensed to deliver both the NVIDIA AI and networking stack—including CUDA Toolkit and DOCA OFED—as a fully validated, ready-to-run solution.

According to CIQ, these expanded options respond to data center and high-performance computing deployment challenges, specifically in scaling from development setups to production environments with thousands of nodes. Now, RLC and RLC-AI use pre-built, validated images that bundle the complete NVIDIA software stack. This includes CUDA Toolkit for GPU acceleration, DOCA OFED for optimized network drivers, and support for remote direct memory access (RDMA), Intel Data Plane Function (IDPF), and secure boot. These features target large-scale GPU clusters, often using servers from major vendors such as Dell Technologies and Hewlett Packard Enterprise (HPE).

CIQ claims that organizations can deploy production-ready environments nine times faster—reducing setup from 30 minutes to three minutes—when transitioning from installation to inference work, based on internal benchmarking with data center-class GPUs and large language model workloads. The stack is designed to optimize GPU-to-GPU communication and minimize both configuration time and troubleshooting across clusters that can span thousands of nodes.

Security features include fully signed GPU drivers and secure boot support. CIQ highlights that RLC and RLC-AI are validated on premium enterprise hardware, addressing GPU driver deployment in modern security-conscious data center environments. The company reports zero-configuration networking for advanced workloads and claims effective elimination of installation guesswork for both developers and operations teams.

Gregory Kurtzer, Founder and CEO of CIQ, stated, “If you’re building applications that leverage accelerated computing, Rocky Linux from CIQ is now the obvious choice,” adding, “We’ve removed every barrier between developers and GPU performance. With the complete, validated NVIDIA stack integrated directly into Rocky Linux from CIQ, teams can focus entirely on innovation rather than infrastructure. This is the appliance experience that the developer community has needed: download, deploy, and start building immediately.”

CIQ will present its updated Rocky Linux solutions and partner reference kits at KubeCon + CloudNativeCon North America (November 10-13, 2025, Atlanta) and SC25 (November 16-21, 2025, St. Louis), featuring deployments with NVIDIA AI infrastructure, NVIDIA ConnectX SuperNICs, and NVIDIA BlueField data processing units (DPUs).

Source: CIQ

Get Data Center Engineering News In Your Inbox:

Popular Posts:

695fcac850f073b041e711a2_karman-p-3200 copy
Karman launches 10 MW Heat Processing Unit for giga-scale AI data center cooling
Screenshot
Five AI data centers to reach 1 GW power capacity in 2026, new analysis shows
1600x1600_1
DCX announces 8.15 MW facility-scale CDU for 45 C warm-water AI data center cooling
Grafika3-scaled copy
DCX announces 8.15 MW coolant distribution unit for 45°C warm-water cooling in AI data centers
Multiple_Stack_with_Calipe_with_Light_Streak_2
Wolfspeed produces first 300mm silicon carbide wafer to boost data center power and cooling efficiency for AI servers

Share Your Data Center Engineering News

Do you have a new product announcement, webinar, whitepaper, or article topic? 

Get Data Center Engineering News In Your Inbox: