An in-depth look at how distributed data, AI readiness, and next-generation protection strategies are reshaping enterprise infrastructure
The Data Revolution Nobody Saw Coming
On my week long journey during the 64th edition of The IT Press Tour in New York, something became crystal clear that most analysts aren’t talking about yet. While everyone’s been focused on AI hardware and cloud migrations, there’s a more fundamental shift happening in how enterprises manage, protect, and extract value from their data. The seven companies we met with – HYCU, Arcitecta, ExaGrid, TextQL, Shade, CTERA, and AuriStor – represent different pieces of this puzzle, but when you step back, they’re all solving variations of the same core problem.
The traditional approach to enterprise data management is breaking down. Hard.
The Distributed Data Protection Challenge
Next-Generation Backup and Recovery
Three of the seven vendors – HYCU, ExaGrid, and Arcitecta – are reimagining data protection for an era where workloads span multiple clouds, edge locations, and hybrid environments. But their approaches reveal how dramatically the backup landscape is evolving.
HYCU positioned itself as an “intelligent data mover” that can protect everything from traditional VMs to AI workloads to SaaS applications. Their Dell partnership, which gives them access to 60,000 channel partners globally, validates their technical architecture while addressing a market reality: backup application vendors are getting out of the hardware business. When customers can protect Microsoft 365, Salesforce, and AI training datasets from a single console while storing data in their own cloud accounts, the traditional backup model starts looking antiquated.
The SaaS protection story particularly resonates. With organizations running an average of 139 SaaS applications (160 for those that experienced breaches), the old approach of managing separate backup tools for each application simply doesn’t scale. HYCU’s expansion to 25 new SaaS integrations in just over a year suggests the market is ready for platform consolidation.
ExaGrid took a different angle, focusing specifically on backup storage with their tiered architecture. Their 19 consecutive quarters of positive cash flow and 99.1% customer support renewal rate speak to solving a real problem: the expanding backup window that plagues traditional scale-up storage. With their non-network-facing repository tier, ExaGrid addresses ransomware protection at an architectural level rather than as an add-on feature.
The company’s AI-powered Auto Detect & Guard feature represents genuine innovation – using pattern recognition to identify anomalous delete patterns and automatically extending retention policies during potential attacks. For MSPs and enterprises dealing with constant ransomware threats, this isn’t just a nice-to-have; it’s business-critical functionality.
Arcitecta brought forward an interesting perspective on long-term data management. Their customer at Princeton University planning a 100-year data retention strategy illustrates a problem most vendors ignore: how do you ensure data accessibility across multiple technology refresh cycles? Their Mediaflux platform, built around an XML-encoded object database, manages over a trillion objects in a single namespace for one customer.
What sets Arcitecta apart is their focus on metadata-driven data management and vendor agnosticism. When Princeton needed to break free from IBM lock-in, they used Mediaflux to unify Dell PowerScale, Dell ECS, and IBM Diamondback libraries under a single management layer. That’s the kind of infrastructure flexibility enterprises need as they navigate vendor relationships over decade-spanning data lifecycles.
Unified Data Access and Management
Breaking Down Data Silos
The second category – represented by TextQL, Shade, and CTERA – addresses data fragmentation from different angles but with a common goal: making distributed data accessible and actionable.
TextQL attacked the enterprise data platform lock-in problem head-on. Their “Rosetta Stone” approach to data integration resonates because it addresses a financial reality: when Dell or SAP migrations cost $50 million and only move 10% of business logic, enterprises need alternatives. As AI adoption drives data platform costs from 10% to potentially 50% of IT budgets, the 20x markups traditional platforms charge become unsustainable.
The company’s pay-as-you-go computing model represents a fundamental shift from CPU-hour billing to compute-time-in-transit billing. For organizations doing real-time analysis across massive datasets, this pricing model makes dynamic workloads economically viable.
Shade solved a different but related problem: creative workflow data fragmentation. When creative teams spend 15 hours weekly managing files across five or more platforms, productivity suffers and costs explode. Their all-in-one platform combining streaming file access, review and approval, and AI-powered organization addresses tool sprawl that affects every organization producing video, graphics, and marketing content.
The semantic search capability – finding “people running along the lake” across thousands of assets – demonstrates how AI can enhance productivity without requiring wholesale workflow changes. With customer cost savings of 55-70% through platform consolidation, Shade’s value proposition resonates with both creative teams and CFO offices.
CTERA presented the most comprehensive vision with their three-wave innovation strategy. Wave 1 (location intelligence) unified file silos across edge, core, and cloud. Wave 2 (metadata intelligence) added real-time ransomware detection. Wave 3 (enterprise intelligence) makes data AI-ready through content analysis and curation.
Their integration with Model Context Protocol (MCP) positions them for the emerging standard in AI-data interaction. The ability to deploy fully private AI workloads while maintaining data sovereignty addresses real concerns killing AI projects in regulated industries.
The AI Data Readiness Problem
Quality Over Quantity
Perhaps the most significant insight from the Tour was how these vendors are approaching AI data preparation. While most of the industry focuses on compute infrastructure and model training, the real bottleneck is data quality.
CTERA’s statistic that 95% of GenAI pilots fail because of poor data quality puts the challenge in perspective. Organizations are trying to “point AI tools at all their data” and wondering why they get “very confident errors” instead of useful insights.
The solution isn’t more storage or faster networks – it’s intelligent data curation. CTERA’s three quality killers (messy data, data silos, and compliance concerns) apply across industries. Their approach of creating curated datasets through format unification, metadata enrichment, and permission-aware access addresses the fundamental challenge of making enterprise data AI-ready.
Arcitecta’s vector database integration shows a clear and measured approach. Rather than building AI capabilities from scratch, they integrate with specialized services for facial recognition, object detection, and OCR, then store vector embeddings alongside traditional metadata. This allows organizations to make data AI-ready without betting their infrastructure on untested AI platforms.
Security and Sovereignty Considerations
Protection as Architecture, Not Add-On
The security theme that emerged across all seven vendors reflects the reality that ransomware isn’t a theoretical threat – it’s a constant operational concern. But the approaches varied significantly.
ExaGrid’s non-network-facing repository represents architectural security. When threat actors can’t see or access the data tier containing retention copies, delete commands only affect the performance tier. Combined with delayed delete policies and AI-powered anomaly detection, this creates multiple layers of protection.
HYCU’s R-Shield integrates security capabilities into the core platform rather than selling them as add-ons. Near-source scanning using existing snapshots avoids the performance penalty of moving petabytes just for malware detection. For customers with data sovereignty requirements, keeping analysis within their environment rather than shipping metadata to cloud services matters.
CTERA’s air-gapped audit trail through Insight provides forensic capabilities without exposing production data. When organizations face insurance companies and auditors after an incident, having detailed records of what was accessed becomes critical for damage assessment and compliance reporting.
The Hidden Innovation: Global File System Evolution
While most vendors at the Tour focused on data protection and AI readiness, AuriStor presented a fundamentally different but equally crucial piece of the enterprise data puzzle – a next-generation distributed file system that’s quietly powering some of the world’s most demanding environments.
Founded in 2007 as Your File System, Inc., AuriStor emerged from the high-energy physics computing community with a mission to modernize distributed file system technology. As Gerry Seidman, President & Member of the Board noted during the briefing: “When hybrid and cloud were the words everyone scrambled to say they had, all we had to do was put it on our marketing material. The product is fundamentally designed to be distributed across different servers, different campuses – it just naturally works in a hybrid multi-cloud environment.”
Technical Innovation at Scale
What makes AuriStor particularly interesting is their focus on solving fundamental distributed computing challenges. Their AuriStorFS system supports:
- Performance up to 8.2 Gbit/second per listener thread (up to 16 threads per service)
- Management of up to 2^64 volume IDs per cell
- Objects up to 16 exabytes in size
- Advanced security with combined identity authentication and AES-256 encryption
The real-world impact is significant. One global financial institution uses AuriStorFS across approximately 80 cells and 300 servers, supporting 175,000+ clients and over 1.5 million volumes. As Jeffrey Altman, Founder and CEO explained: “Production AuriStorFS fileservers regularly service more than half a million client connections simultaneously.”
Container-Ready Architecture
Their container integration story is particularly relevant for modern enterprises. AuriStor has developed a Container Storage Interface (CSI) driver that integrates directly with Red Hat OpenShift, addressing a critical challenge in container deployments. For machine learning workloads where container images often exceed 40GB but only 5% is accessed during execution, AuriStor’s implementation allows containers to mount AFS volumes directly, fetching data on demand.
This capability becomes crucial for organizations transitioning to containers at scale. One financial customer is using this approach to migrate 180,000 systems to containers while maintaining their existing software distribution workflow – handling 400GB distribution trees that update every few minutes.
A Different Pricing Model
In an industry obsessed with storage-based pricing, AuriStor takes a refreshingly different approach. As Seidman humorously noted, “We give you the first 100 exabytes for free!” Their pricing model focuses on servers and user identities rather than storage capacity, making costs more predictable for enterprises.
While AuriStor might not fit neatly into the AI and data protection narratives that dominated the Tour, their technology represents a crucial foundation layer for enterprises building modern distributed applications. Their success in financial services, research institutions, and government agencies demonstrates that innovation in core infrastructure remains vital, even as attention shifts to AI and analytics.
Market Dynamics and Industry Implications
The Great Disaggregation
Several trends emerged that will reshape the enterprise data infrastructure market:
Backup Application Vendors Abandoning Hardware: The shift toward subscription revenue models and SaaS offerings doesn’t align with hardware sales. As Commvault, Veritas, and others focus on software, they’re creating opportunities for specialized storage vendors like ExaGrid.
Platform Consolidation: Organizations are tired of managing separate tools for backup, archive, collaboration, and analytics. Vendors that can unify these capabilities under consistent management interfaces have significant advantages.
Vendor Agnosticism: The days of single-vendor infrastructure stacks are ending. Organizations want the flexibility to choose best-of-breed solutions and change vendors without massive migration projects. Platforms that support bring-your-own-storage and open standards have advantages.
AI as a Forcing Function: AI workload requirements are driving infrastructure decisions. Organizations need platforms that can make their data AI-ready without requiring complete infrastructure replacement.
The Emerging Data Intelligence Stack
Beyond Traditional Storage
Looking across all seven vendors, the outlines of a next-generation data intelligence stack become visible:
Foundation Layer: Global namespace and unified access (CTERA, Arcitecta, Shade)
Protection Layer: Real-time threat detection and immutable retention (ExaGrid, HYCU)
Intelligence Layer: AI-powered data curation and semantic search (TextQL, CTERA, Shade)
Integration Layer: Open standards and vendor-agnostic connectivity (MCP, FUSE, S3)
This isn’t about replacing existing infrastructure overnight. It’s about adding intelligence and unification layers that make distributed data more accessible, better protected, and ready for AI workloads.
Looking Forward: What Enterprises Should Watch
Strategic Implications
Based on the technologies and approaches showcased at the Tour, several strategic considerations emerge for enterprise IT leaders:
Data Sovereignty Will Drive Architecture Decisions: As AI adoption accelerates, organizations need platforms that can process data locally while accessing global AI capabilities. Hybrid deployment models that keep sensitive data on-premises while leveraging cloud AI services for inference will become standard.
Protection Strategies Must Evolve: Traditional backup-and-recovery approaches can’t address modern ransomware tactics. Organizations need platforms that integrate protection, detection, and forensics capabilities rather than managing separate point solutions.
Vendor Lock-in Costs Will Become Unsustainable: As data platform costs potentially grow from 10% to 50% of IT budgets, the markup penalties of proprietary platforms will drive adoption of open, interoperable solutions.
AI Success Requires Data Curation: The organizations that succeed with AI will be those that invest in data quality and curation rather than just throwing larger models at messy datasets.
The Road Ahead
The 64th IT Press Tour revealed an industry in transition. While the headlines focus on AI chips and model training, the real innovation is happening in data infrastructure. The vendors we met represent different aspects of this evolution, but they share a common vision: making enterprise data more accessible, better protected, and ready for intelligent analysis.
The winners in this space won’t be the vendors with the flashiest AI features or the largest marketing budgets. They’ll be the ones that solve fundamental data management problems while positioning organizations for the next decade of AI-driven business transformation.
For enterprise IT leaders, the message is clear: the infrastructure decisions you make today about data management, protection, and access will determine your organization’s ability to compete in an AI-driven future. Choose platforms that provide flexibility, intelligence, and vendor independence. The alternative is getting locked into expensive, proprietary systems that become anchors rather than enablers.
The data revolution is here. The question isn’t whether your organization will be affected – it’s whether you’ll be prepared.
##
The 64th IT Press Tour brought together technology journalists and industry leaders to examine the latest innovations in enterprise data infrastructure. The insights shared by HYCU, Arcitecta, ExaGrid, TextQL, Shade, CTERA, and AuriStor provide a comprehensive view of how the industry is evolving to meet the challenges of distributed data, AI readiness, and next-generation security threats.
