Virtana announced the launch of Virtana AI Factory Observability (AIFO), a new capability that extends Virtana’s full-stack observability platform to the unique demands of AI infrastructure.
With deep, real-time insights into everything from GPU utilization and training bottlenecks to power consumption and cost drivers, AIFO enables enterprises to turn complex, compute-intensive AI environments into scalable, efficient, and accountable operations.
“AI has the potential to be as transformative as the steam engine or the printing press—but only if enterprises can operationalize it at scale,” said Paul Appleby, CEO of Virtana. “Right now, too many teams are flying blind when it comes to AI infrastructure. Virtana AIFO gives them the visibility and control they need to treat AI not as an experiment, but as a core, strategic part of the business.”
Virtana AIFO helps enterprises treat AI infrastructure with the same level of visibility, discipline, and accountability as traditional IT.
As an official NVIDIA partner, Virtana integrates natively with NVIDIA GPU platforms to deliver in-depth telemetry, including memory utilization, thermal behavior, and power metrics, providing precise, vendor-validated insight into the most performance-critical components of the AI Factory. This deep integration delivers accurate, actionable intelligence at enterprise scale.
“AI workloads introduce an entirely different set of infrastructure challenges—from GPU saturation and training bottlenecks to unpredictable cost spikes,” said Amitkumar Rathi, Senior Vice President of Engineering, Product, and Support at Virtana. “We designed AIFO to address these realities head-on. It gives teams deep, correlated visibility across the full AI stack, enabling them to optimize performance, reduce waste, and scale AI with confidence.”
With this launch, Virtana directly addresses the growing infrastructure challenges that stand in the way of scalable AI success. As enterprises accelerate investments in AI, many are encountering hidden inefficiencies: idle GPUs that inflate costs, training jobs that fail without explanation, and inference pipelines that stall due to underlying storage or network issues. AIFO is purpose-built to solve these problems, delivering real-time visibility and correlated insights across every layer of the AI infrastructure stack. The result is greater control over performance, spend, and scale—turning AI from a high-risk initiative into a high-impact capability.
Virtana AIFO is purpose-built to meet the demands of AI operations. It continuously collects telemetry across GPUs, CPUs, memory, network, and storage and then correlates that data with training and inference pipelines to provide clear and actionable insights.
Core capabilities include:
- GPU Performance Monitoring – Tracks per-GPU metrics such as memory, utilization, thermal load, and power draw across multiple vendors.
- Distributed Training Visibility – Identifies bottlenecks, synchronization issues, and stragglers across multi-node jobs.
- Infrastructure-to-AI Mapping – Correlates model-level performance directly to hardware-level behavior, including network and storage dependencies.
- Power and Cost Analytics – Exposes inefficiencies such as thermal throttling, idle GPU time, and overprovisioning resources.
- Root Cause Analysis – Diagnoses training failures and inference slowdowns faster by pinpointing the most likely infrastructure causes.
All capabilities are accessible via Virtana’s Global View dashboard, which unifies telemetry across hybrid and containerized AI environments—on-premises, cloud, or both.
AIFO is already delivering measurable results in production AI environments across multiple industries. Operational outcomes include:
- 40% reduction in idle GPU time, improving resource utilization and reducing infrastructure costs.
- 60% faster mean time to resolution (MTTR) for AI-related incidents.
- 50% decrease in false alerts, reducing operational noise and accelerating response.
- 15% improvement in power efficiency, supporting sustainability goals.
Virtana AIFO is now generally available as a fully integrated capability within the Virtana Platform. Purpose-built for the demands of modern AI infrastructure, AIFO scales effortlessly from early-stage test environments to enterprise-grade AI factories. This launch, together with Virtana’s recent acquisition of Zenoss, further extends the company’s leadership in delivering the deepest, and broadest observability platform across applications, infrastructure, and AI workloads in hybrid and multi-cloud environments.
Additionally, Virtana’s recent acquisition of Zenoss expands the platform’s event intelligence and service-centric observability capabilities, allowing customers to correlate AI model performance with broader application behavior and infrastructure health. Together, these advancements deepen Virtana’s ability to help enterprises manage the full complexity of AI operations in the most demanding environments.
The Latest
I've spent a lot of time in the channel, and one thing I keep coming back to is this: a partner program is only as good as what it looks like in the field. Many programs look great on paper, but when a partner is in front of a customer navigating a complex hybrid environment or trying to make the case for AI-powered observability, the gap between what a vendor promises and what it actually delivers becomes very clear, very fast ...
Enterprises today operate in a real-time environment where uninterrupted access to trusted data has become a baseline expectation for users, applications and automated systems. Traditional DataOps models, built on manual effort and human triage, cannot keep pace with this always active demand. AI agents are emerging as the operational backbone, ensuring consistent data availability, reinforcing trustworthiness and enabling a level of scale that manual processes cannot achieve ...
For decades, trust in the digital workplace rested on familiar signals. We trusted faces on video calls, voices on the phone, and emails that appeared to come from people we knew. These cues felt human and intuitive. They anchored how decisions were made, approvals were granted, and access was authorized. AI-powered deepfakes have quietly broken that model ...
Cloud migration was supposed to be a one-way door. For most enterprises, it turns out it isn't. Cloud data repatriation is a real and growing trend. A new survey ... finds that 89% of organizations plan to expand their on-premises infrastructure footprint over the next two years — and 75% have already moved at least some workloads back from public cloud in the past 24 months. The findings point to a broad rethinking of where data belongs ...
Over the past few years, large language models (LLMs) have revolutionized the software industry. Given their ability to excel at multi-step reasoning, LLMs have helped enterprises streamline workflows and adapt to the unknown. However, employing such models comes with sky-high costs, latency issues, and limited flexibility. In the realm of IT operations, it is generally wiser to employ smaller, domain-specific models instead ...
For years, DevOps teams operated under a simple assumption: collect enough telemetry, and you can find and fix any problem. That assumption is breaking down. Modern enterprises now operate across microservices, hybrid cloud environments, APIs, Kubernetes, and highly automated delivery pipelines. Releases happen continuously, dependencies shift constantly, and failures spread faster than teams can diagnose them ...
New Relic surveyed IT and engineering leaders from the media and entertainment (M&E) sector to understand what's working — and where challenges persist with their observability practices. The findings reveal how M&E organizations are navigating rising platform complexity, audience expectations, and AI-driven change. Below are five takeaways that stand out ...
Let me start with something I've seen play out more times than I can count. A team hits a wall with the cloud. Costs creep up, then spike. Performance starts to feel inconsistent. Someone in finance asks a simple question like "why did this double?" and nobody has a clean answer ... Maybe this isn't the right place for everything. That realization feels like a breakthrough, like you've identified the problem. In reality, you've just identified the starting line ...
In MEAN TIME TO INSIGHT Episode 24, Shamus McGillicuddy, VP of Research, Network Infrastructure and Operations, at EMA discusses network observability tool sprawl ...
In cloud-native systems, scaling is often as simple as moving a slider. For on-premise databases, the stakes are different. Over-provisioning hardware is expensive. Under-provisioning leads to performance bottlenecks that are difficult to fix once the equipment is in the rack ...