CIQ

The fear that stops most RHEL migrations isn't the cost of switching. It's the operational risk. Rebuilding hundreds or thousands of servers is expensive in both time and talent. Re-testing applications, rewriting automation, compliance recertification adds up fast. Teams stay put, even when they want a change. With RLC Pro migrating from RHEL doesn't require any of that. Full binary compatibility means your applications, deployment scripts, and Ansible playbooks carry over. The switching cost is deployment time, not re-engineering time. We just published a step-by-step guide walking through the full process: from pre-migration assessment to production cutover. Read the guide: bit.ly/40BPbcl #EnterpriseLinux #RockyLinux #RLCPro #RHEL #Migration #OpenSource

Tokens per watt is the new CEO metric, and most infrastructure teams are leaving it on the table at the OS layer. Unvalidated drivers. Manual tuning cycles. PyTorch configs that were never tested against the rest of the stack. All throughput ceilings the hardware never created. RLC Pro AI reduces OS tuning and update cycles from hours to minutes. The stack ships validated. Nothing needs to be assembled. Tokens per watt goes up before you run a single workload. ciq.com/blog/tokens-pe… #NVIDIAGTC #GTC2026 #EnterpriseLinux @NVIDIA

The CIQ portal is live: access, evaluate, and deploy CIQ products on your own terms. From registration to your first download in under 10 minutes. Personal Use, Free, Developer and Trial licenses are available today, alongside a full product catalog including RLC Pro, RLC Pro Hardened, RLC Pro AI, RLC+, CIQ Bridge, Fuzzball, Ascender Pro, and Warewulf Pro. Start where you are. Scale when you're ready: bit.ly/4bRs2Jc #RLCPro #RockyLinux #Linux #OpenSource #EnterpriseLinux #CIQ

GROMACS, AlphaFold, NVIDIA Inference, and Jupyter notebooks. All on the same platform. Watch @David Godlove demo how Fuzzball unifies HPC and AI workloads on a single infrastructure, with no manual resource wrangling required: youtu.be/-1E30PTAQ7c #HPC #AI #MLOps #Fuzzball

@drushtx Good looking shirt!

@CtrlIQ I love Rocky on St. Patrick's Day!

If you've spent an afternoon chasing a CUDA dependency conflict, this one's for you. On April 2, join Damen Knight, Zach Deane-Mayer, Brian Dawson, and Eric Hendricks as we walk through why the OS underneath your GPU fleet is where AI performance is won or lost, and what it looks like when that layer is actually built for the workload. Maximize the Throughput of Your AI Infrastructure April 2, 2026 | 2:00 PM ET | 60 minutes Live deployment walkthrough included. Q&A at the end. Register: bit.ly/46TMpmd #AIInfrastructure #EnterpriseLinux #NVIDIA #CUDA #RLCPro #GPU #RLCProAI

Jensen at #NVIDIAGTC: inference has overtaken training as the dominant AI workload. The OS underneath your GPUs is now part of the performance equation. RLC Pro AI: up to 32% faster vision throughput and 10% faster LLM inference on identical hardware. Pre-validated NVIDIA CUDA + DOCA-OFED. Ready at first boot. The inference era is here. We built the OS for it. ciq.com/blog/nvidia-ju… #GTC2026 #ShowMeGTC #AIInfrastructure @NVIDI

Already running Slurm or PBS? Don't rip it out. Build on it. Watch David Godlove show how Fuzzball integrates with your existing HPC infrastructure, GPUs, cloud bursting, Nextflow, and more, without disrupting a single existing workflow: youtu.be/ZEp8URhkG58 #HPC #AI #Slurm #Fuzzball

We made it to Pasadena for @socallinuxexpo 23x, and it lived up to every bit of the hype. Our team spent the week on the expo floor having real conversations with the open source community, from longtime Rocky Linux users to folks just discovering what CIQ does. @elgueromexicano delivered a hands-on hardening workshop and a talk on Secure Boot. @itguyeric got honest about burnout in tech. We shared donuts with our neighbors, got humbled at Uno No Mercy during Game Night, and topped 300 booth badge scans. We had a great time and came home with a lot to follow up on. If you missed us there, our full recap is on the blog: bit.ly/4bhGv0M #SCaLE #OpenSource #RockyLinux #Linux #CIQ

HPC and AI workloads are converging. Is your infrastructure ready? Watch David Godlove break down what Fuzzball is, why it exists, and why neither traditional HPC schedulers nor Kubernetes are cutting it for modern research teams: youtu.be/tc4Z1pewr-c #HPC #AI #CloudComputing #Fuzzball

Did you miss our webinar last week on RLC Pro? Brady Dibble and @itguyeric spent an hour breaking down the real differences between Community Rocky Linux, RLC+, and RLC Pro, and what it actually means to have a vendor stand behind your enterprise Linux infrastructure. Topics covered: Long-Term Support, FIPS 140-3 validation, direct bug fixes, and how organizations are moving to RLC Pro without re-architecting their existing workloads. The recording is now available. Watch it here: youtu.be/_uDtPH1Bay4 #EnterpriseLinux #RockyLinux #RLCPro #LTS #FIPS

One command. One shell. One GPU job. That's all it takes with Fuzzball v3.2. Read Jonathon Anderson's full breakdown of what's new: real-time workflow events and self-service password management: bit.ly/4cM2vCd #HPC #AI #CloudComputing #Fuzzball

.@fossforce is covering our RLC Pro AI launch, and they put it well: "the OS is ground zero, or where the stack meets the metal." Read their take on what RLC Pro AI delivers and why the OS layer matters for GPU infrastructure. bit.ly/3P7zoiP #AIInfrastructure #EnterpriseLinux #RLCPro #GPU

Your GPU fleet is already paid for. The hardware can do more than it's delivering. The question is how much of its capacity you're actually getting. The OS underneath most GPU clusters was never built for AI/ML workloads. RLC Pro AI: The Enterprise Linux built to maximize your AI infrastructure. RLC Pro AI is generally available! RLC Pro AI ships with the CIQ Linux Kernel, the full NVIDIA CUDA and DOCA-OFED stack, and PyTorch, — pre-validated and pre-configured for AI inference from first boot. No manual CUDA assembly. No waiting for the OS to catch up to your hardware. Same validated stack across bare metal, AWS, GCP, Azure, and on-premises. More output from the GPUs you already own. Read the full post from President of CIQ, @bjorn hovland: bit.ly/3NBsCkZ #AIInfrastructure #EnterpriseLinux #NVIDIA #CUDA #RLCPro #GPU

RLC Pro AI is generally available! Most enterprises are running GPU infrastructure on operating systems that were never built for AI/ML workloads. The OS layer is where performance is won or lost, and for most organizations, that performance has been sitting on the table. RLC Pro AI: The Enterprise Linux built to maximize your AI infrastructure RLC Pro AI ships with the CIQ Linux Kernel, the full NVIDIA CUDA and DOCA-OFED stack, and PyTorch, pre-validated and pre-configured for AI inference from first boot. No manual CUDA assembly. No waiting for the OS to catch up to your hardware. The same validated stack across bare metal, AWS, GCP, Azure, and on-premises. More output from the GPUs you already own. Available now. Read the full announcement: bit.ly/4sKvJpS #AIInfrastructure #EnterpriseLinux #NVIDIA #CUDA #RLCPro #GPU

1 hour until we go live! Brady Dibble and @itguyeric are breaking down the real differences between Community Rocky Linux, RLC+, and RLC Pro and what it actually means to have a vendor stand behind your enterprise Linux. Join us at 11am PT | 2pm ET: bit.ly/4l09MAk #EnterpriseLinux #RLCPro

@itguyeric It's not to late to register!

⏰ 24 hours to go Tomorrow we’re breaking down one of the most confusing questions in #EnterpriseLinux: When do you actually need a commercial Linux subscription? Join Brady Dibble and me tomorrow. Register: bit.ly/4l09MAk #RockyLinux #Linux #RLCPro @CtrlIQ

Five idle drives on a compute node = wasted scratch performance. Learn how to auto-assemble them into a software RAID array with Warewulf, mdadm, and Ignition. No manual per-node work required. Full guide: bit.ly/47tkbin

@tomvideo2brain Great explanation.

The Active Cyber Program (ACP) is a cybersecurity assessment and certification framework designed to evaluate whether an organization operates an active and effectively managed cybersecurity program. @MVPAward @ISC2 @NIST @msftsecurity @CtrlIQ linkedin.com/pulse/active-c…

@elgueromexicano We are so glad you went!

It was a wonderful time there with our OSS family. #fosdem #oss