Protecting Enterprise Data at the Source with NeuralProtect™ 

Cyber resilience remains one of the most urgent priorities for every enterprise IT and security leader. Ransomware attacks continue to evolve rapidly, targeting the most critical enterprise asset of all: production data.

Traditional ransomware protection strategies often focus primarily on endpoint detection or backup recovery. But by the time encryption is detected, the damage may already be done.

That’s why we built Qumulo NeuralProtect™.

NeuralProtect™ delivers AI-driven, real-time ransomware protection directly at the storage layer, inspecting files as they are written and identifying threats before enterprise data is encrypted, corrupted, or lost.

This fundamentally changes ransomware defense from a reactive recovery model into a proactive prevention model.

The best place to stop ransomware is where the attack ultimately targets: the data itself.

NeuralProtect™ uses multiple AI detection approaches to identify both known and emerging threats, including deterministic AI models for known ransomware variants, statistical AI models for zero-day attacks, and temporal AI models for stealth attacks and partial encryption.

Unlike entropy-based approaches that rely primarily on metadata or behavioral assumptions, NeuralProtect™ performs a deep inspection of actual file contents at the point of write. That enables earlier detection, reduced false positives, and protection against more sophisticated attack techniques.

But enterprise security today requires more than isolated detection.

That’s where our partnership with Cisco becomes incredibly powerful.

By integrating NeuralProtect™ with Cisco Hypershield and Splunk, Qumulo extends ransomware defense beyond storage to coordinated, enterprise-wide threat containment and response.

Together, Qumulo and Cisco enable organizations to:

• Detect ransomware in real time
• Automatically isolate compromised systems
• Trigger snapshots and quarantine actions
• Extend enforcement across infrastructure and networks
• Improve visibility through integrated telemetry and observability

Cisco Hypershield helps orchestrate network-level containment, while Splunk integration provides enterprise-wide observability and incident-response workflows.

The result is a unified cyber resilience architecture spanning storage, infrastructure, and security operations.

For enterprises operating across hybrid cloud and distributed environments, that level of coordinated protection matters more than ever.

Building a Practical Bridge to the Cloud

Hybrid infrastructure has become the operational reality for most enterprises. But many organizations still face a difficult challenge: how to modernize infrastructure and extend workloads into the cloud without disrupting mission-critical operations.

Too often, cloud modernization projects become massive migration exercises that introduce operational risk, application rewrites, user disruption, and years-long timelines.

We believe there’s a better approach.

At Cisco Live, Qumulo is introducing the Qumulo CNQ Enterprise + Cisco UCS Bridge-to-Cloud solution, designed to help enterprises extend high-value file workloads into cloud environments without disruptive migration or application refactoring.

This launch combines the strengths of Qumulo’s hybrid cloud data platform with Cisco’s trusted enterprise infrastructure foundation.

CNQ Enterprise brings together:

• Cloud Native Qumulo
• Unlimited Cloud Data Fabric
• Qumulo NeuralProtect™ 

into a unified hybrid cloud platform available across AWS, Azure, Google Cloud, and Oracle Cloud Infrastructure.

Cisco UCS and Cisco networking provide the scalable compute, integrated networking, operational consistency, and secure connectivity required for enterprise hybrid cloud deployments.

Together, Qumulo and Cisco help enterprises modernize on their own timeline while preserving operational continuity. CNQ Enterprise is available directly through Cisco for simplified enterprise procurement. 

One of the most important capabilities here is Cloud Data Fabric.

Cloud Data Fabric allows distributed enterprise datasets to appear local regardless of where the data physically resides. Applications and users maintain seamless access while organizations intelligently extend workloads across on-premises and cloud environments.

That means enterprises can:

• Extend storage capacity instantly
• Reduce pressure on expensive hardware refresh cycles
• Avoid costly replatforming engagements
• Extend to the cloud without application or user access disruption
• Preserve existing workflows and user experiences

In today’s constrained hardware market, that flexibility becomes strategically important.

Building a bridge to the cloud gives enterprises time, optionality, and operational freedom.

It also creates the foundation for enterprise AI.

As organizations increasingly deploy AI and analytics workloads across distributed environments, they need infrastructure that can scale elastically while supporting global collaboration and data-intensive operations. CNQ Enterprise was designed specifically for that future.

Unlocking True “GPU Liquidity” in Enterprise AI

Every enterprise pursuing AI initiatives eventually runs into the same issue: GPU availability, accessibility, and utilization.

Finding available GPUs has become a constant operational challenge, especially across regions, cloud providers, and hybrid infrastructure environments. But increasingly, the industry is realizing the bigger problem is not simply GPU availability. Its utilization.

A recent analysis highlighted a staggering reality facing enterprise AI initiatives: average enterprise GPU utilization hovers around 5%.

That means hundreds of billions of dollars’ worth of accelerated compute infrastructure sits idle roughly 95% of the time.

The industry response has often been predictable. Traditional infrastructure vendors continue pushing enterprises toward increasingly expensive, tightly coupled AI storage architectures designed to maximize throughput during active compute windows. The assumption is that if AI projects are stalling, organizations simply need faster storage arrays directly attached to GPU clusters.

But optimizing storage performance during a tiny percentage of active runtime does not solve the larger operational problem.

The true bottleneck in enterprise AI is not raw storage performance. It’s data gravity.

Massive enterprise datasets are difficult, expensive, and time-consuming to move. Traditional AI workflows require organizations to repeatedly replicate or stage data before workloads can begin. Teams spend days or weeks moving datasets into isolated “AI storage islands” attached to dedicated GPU environments. During that entire process, expensive GPU infrastructure often sits idle while data preparation finishes.

At Cisco Live, Qumulo is introducing the Cloud AI Accelerator to solve exactly this problem.

Qumulo Cloud AI Accelerator transforms GPU accessibility from a logistical gamble into a flexible scheduling operation.

Instead of forcing enterprises to pay for idle GPUs as they move massive datasets wherever GPUs are available, Qumulo keeps data accessible through a unified AI data fabric spanning on-premises, edge, and multi-cloud environments.

By leveraging Cloud Native Qumulo, Cloud Data Fabric, and NeuralCache, enterprises can deploy AI Accelerators in any cloud region and present datasets directly to available GPU resources without replicating massive datasets or losing data consistency.

This fundamentally changes the economics and operational model of enterprise AI.

Organizations can:

• Run workloads wherever and whenever GPUs become available
• Eliminate weeks-long data staging delays
• Reduce idle GPU costs by avoiding the load phase into GPU attached flash
• Avoid maintaining multiple replicated storage islands
• Accelerate AI iteration cycles
• Reduce cloud egress and duplication costs

Rather than building isolated performance enclaves, Qumulo enables enterprises to create a unified AI data fabric that delivers real-time access to distributed datasets across clouds and regions.

The architecture is designed for real enterprise AI at scale.

Cloud Native Qumulo scales to more than 2 TB/s and over 20 million IOPS on AWS, while supporting modern AI orchestration frameworks including Kubernetes, Slurm, Ray, and SkyPilot, and also providing zero-copy integrations with Microsoft AI Foundry, AWS Bedrock, and Google Vertex AI. 

Importantly, performance scales independently from storage capacity, allowing organizations to optimize both cost efficiency and AI throughput.

Cisco’s networking, security, and compute play a foundational role here.

Cisco UCS provides scalable enterprise AI compute infrastructure, while Cisco networking enables secure, high-performance connectivity across hybrid and multi-cloud AI environments.

Together, Qumulo and Cisco enable enterprises to build agile AI infrastructure that adapts in minutes to changing GPU availability across clouds and regions.

This is what modern AI infrastructure should look like: flexible, distributed, performant, and operationally simple. This is how you create GPU liquidity through an intelligent data fabric. 

A Shared Vision for the Future of Enterprise Infrastructure

Across all three launches, a common theme emerges.

Enterprises no longer want rigid infrastructure models that force disruptive migrations, siloed operations, or fragmented security strategies.

They want flexibility.

They want operational consistency across environments.

They want AI-ready, empowered, and connected infrastructure.

And they want security embedded directly in the data layer and coordinated with the network layer.

Qumulo and Cisco are building that future together.

Whether organizations are modernizing hybrid cloud infrastructure, protecting critical enterprise data from ransomware, or accelerating AI pipelines across distributed GPU environments, our partnership is focused on helping enterprises move faster without sacrificing control.

At Qumulo, we believe the future belongs to organizations that can securely move data, workloads, and intelligence anywhere they need to operate.

Any Data. Any Location. Total Control.

We’re excited to continue that journey with Cisco at Cisco Live 2026 and beyond.

Experience live demonstrations at Qumulo Booth #4018 at Cisco Live 2026 in Las Vegas (May 31–June 4)

Reference: April 21, 2026: Cast AI’s 2026 State of Kubernetes Optimization Report Reveals GPU Utilization at 5%
https://cast.ai/press-release/2026-state-of-kubernetes-optimization-report/