Remote First Company | VAST Forward | Salt Lake City, Utah, February 25, 2026 (Globe Newswire) — Today VAST Forward 2026, VAST dataan AI operating system company, announced a fully CUDA-accelerated end-to-end AI data stack, delivered through an expanded collaboration with NVIDIA. and VAST AI Operating System Now running directly on NVIDIA-powered servers, customers can eliminate data bottlenecks across their AI pipelines and enable ingestion, acquisition, analysis, and inference on a single unified platform.
By accelerating both data services and compute layers as one coherent system, VAST AI OS eliminates the operational complexity of stitching together separate storage, database, and AI infrastructure stacks. The result is a simpler and faster path from experimentation to production for RAG pipelines, agent systems, and continuous AI workloads.
Designed in collaboration with NVIDIA, VAST CNode-X is a new generation NVIDIA certified system Transform the way you build and operate your AI infrastructure. In addition to providing high-performance storage services to NVIDIA GPU-accelerated clusters, the VAST AI OS now runs directly on NVIDIA-powered servers, making these systems part of the first-class infrastructure within the VAST platform. This architectural change allows VAST to orchestrate AI pipelines, high-performance analytics, vector search, RAG functionality, and agent runtimes as a single integrated software stack.
The new CNode-X server provides the compute foundation for the VAST AI OS and allows you to leverage various NVIDIA software libraries and APIs directly within the core VAST software services. VAST data engine and VAST database. These accelerations are built deep into the platform to deliver higher performance, lower latency, and increased efficiency across real-time SQL analysis, vector search and retrieval, and a wide range of AI inference workflows.
“Ten years ago, we set out to build a system that could continuously refine data and turn it into intelligence and action.” Renen Hallak, Founder and CEO of VAST Data. “The future is here. By accelerating both the compute and data paths within the VAST AI OS with NVIDIA, we are giving our customers a faster and easier way to operate their acquisition, analysis, and agent workflows as one coherent pipeline, enabling AI to move from pilot systems to durable production systems.”
“NVIDIA is reinventing every pillar of computing for AI. With VAST Data, we are transforming storage for AI infrastructure.” Jensen Huang, Founder and CEO of NVIDIA. “CNode-X is CUDA-accelerated at every layer and provides persistent memory for AI agents so they can work on complex problems for days, weeks, and ultimately years without forgetting, opening the world to the next frontier in AI.”
watch this video together NVIDIA Founder and CEO Jensen Huang at VAST Forward of The future of data infrastructure.
Built on the new GPU-accelerated VAST CNode-X Server, VAST integrates extensive support for NVIDIA acceleration features within the VAST AI OS and deploys them within a full-stack software platform that runs and orchestrates AI pipelines, vector search services, and production AI pipelines. New features include:
- GPU-native SQL engine acceleration for VAST database analytics pipelines: VAST is evolving the VAST database to accelerate modern analytical workloads across the query lifecycle by combining storage-side intelligence with GPU-accelerated execution. The VAST DataBase query engine combines intelligent data layout, pushdown, and filtering to reduce unnecessary I/O, while delivering GPU-accelerated SQL execution at the compute layer using Sirius, an open-source query engine based on NVIDIA cuDF. NVIDIA cuDF is a library for accelerating structured data analysis. This complementary approach accelerates both the processing that occurs before the data reaches compute and the compute itself, providing a database that is both storage-optimized and GPU-accelerated. Early benchmarks for Sirius showed query times reduced by up to 44 percent and query costs reduced by up to 80 percent.
- NVIDIA cuVS For fast vector search and retrieval: By incorporating the NVIDIA cuVS library, CNode-X for VAST brings GPU acceleration to vector search and data clustering for organizations using VAST for scalable vector database services. VAST Insight Engineis built on the NVIDIA AI Data Platform reference design for production RAG pipelines to improve acquisition latency for real-time, context-rich AI applications.
- NVIDIA Nemotron Model for a scalable DataEngine pipeline and NVIDIA NIM microservices: VAST is now expanded and supported NVIDIA NIM Microservices Enable a scalable AI pipeline across CNode-X, open source Production-ready VAST DateEngine blueprints for AI pipelines targeting video intelligence, enterprise document RAG, and genomics research use cases.
- Accelerating inference at scale with NVIDIA CMX: VAST support NVIDIA Contextual Memory Storage (CMX) PlatformUse a cluster configuration that supports NVIDIA BlueField-4 DPU and Spectrum-X Ethernet networking speeds access to a shared KV cache and reduces time to first token for long-context multi-agent inference. This gives the agent access to the entire pod’s memory. VAST Disaggregated All Shares (DASE) This architecture also has the added benefit of allowing customers to optionally add out-of-band enterprise data services without sacrificing KV retrieval time.
Choosing Hardware to Accelerate the VAST AI Operating System
VAST plans to bring CNode-X servers to market through major OEM partners such as Cisco and Supermicro, allowing customers to procure GPU-accelerated infrastructure through their preferred vendor while maintaining a consistent VAST software, support, and operational experience.
VAST provides a faster, more supportable path to production AI through certified configurations offered with OEM partners. As enterprise AI pipelines become continuous systems, VAST combines its data platform with full-stack NVIDIA-accelerated computing to deliver high-performance search, analytics, and vector search that maintains GPU productivity across RAGs, real-time analytics, and large-scale AI workloads.
“AI doesn’t scale through isolated components; it scales through integrated systems,” he said. Jeremy Foster, Senior Vice President and General Manager, Cisco Computing. “Customers need an infrastructure that keeps their data safe and works closely with intelligent networking and GPU-accelerated computing to deliver efficient, production-ready platforms. Cisco’s collaboration with partners like VAST and NVIDIA provides the enterprise-ready foundation organizations need to securely scale AI with performance, resiliency, and control.”
“Production AI requires new levels of integration across compute, acceleration, and data platforms,” he said. Charles Liang, Supermicro President and CEO. “We are working with VAST Data and NVIDIA to deliver a truly integrated AI data platform that takes the complexity out of enterprise AI. By bringing together high-performance computing, scalable data infrastructure, and intelligent software in one solution, organizations can quickly move from experiment to production and unlock real business value from AI.”
Additional resources:
About VAST data
VAST Data is an AI operating systems company that powers the next generation of intelligent systems with an integrated software infrastructure stack purpose-built to unlock the full potential of AI. VAST AI OS integrates fundamental data, compute services, and agent execution into one scalable platform, enabling organizations to deploy and facilitate communication between AI agents, reason on real-time data, and automate complex workflows on a global scale. Built on VAST’s groundbreaking DASE architecture, the world’s first truly parallel distributed system architecture that eliminates tradeoffs between performance, scale, simplicity, and resiliency, VAST has transformed modern infrastructure into a global fabric for inferential AI. Learn more here vast data.com Follow VAST data linkedin, YouTube and ×.
