Pure Storage has announced it is integrating the NVIDIA AI Data Platform reference design into its FlashBlade platform, expanding its commitment to deliver validated, enterprise-grade scalable, AI-ready solutions for customers that meet NVIDIA’s rigorous standards.
Pure Storage is advancing enterprise data storage and management by enhancing enterprise customers’ ability to manage the high-performance data and accelerated compute requirements necessary for successful AI deployments, including:
- Delivering NVIDIA AI Data Platform capabilities with Pure Storage FlashBlade: Supporting NVIDIA’s new reference design for building intelligent AI Data Platforms, FlashBlade enables customers to elevate storage in the era of AI, as a high-performance, distributed system that rapidly provides business insights from data. By tapping into NVIDIA accelerated computing, networking, and NVIDIA AI Enterprise software, FlashBlade can process data with the speed and agility required for AI reasoning applications.
This announcement follows the launch of Pure Storage’s FlashBlade//EXA, the industry’s highest-performing data storage platform engineered for the most demanding requirements of AI and high-performance computing.
- Achieving certification as a high-performance storage (HPS) platform for NVIDIA Partner Network Cloud Partners: Pure Storage is now officially HPS certified for NVIDIA Cloud Partner Reference Architectures with NVIDIA HGX systems using NVIDIA B200 or H200 GPUs, further strengthening our collaboration with NVIDIA.This certification acknowledges Pure Storage as a trusted storage partner and validates Pure Storage’s ability to support NVIDIA Cloud Partners with a proven reference architecture that meets all performance and technical requirements for building state-of-the-art GPU clouds.
- Achieving NVIDIA-Certified Storage Partner approval for both Foundation and Enterprise levels, showcasing capabilities for enterprise AI factories: NVIDIA-Certified Storage Systems addresses the massive data demands of AI factories by offering a comprehensive storage certification. It rigorously tests storage systems to ensure they meet the highest standards of quality, efficiency, and reliability. This programme complements NVIDIA-Certified Systems and NVIDIA Enterprise Reference Architectures (RAs), enabling partners and customers to efficiently deploy AI infrastructures. By integrating validated storage with NVIDIA’s accelerated computing and software technologies, customers can confidently build AI factories that leverage data for faster, more accurate, and reliable AI models, unlocking new possibilities and driving innovation across industries.
Along with Pure Storage’s recent certification of FlashBlade//S500 with NVIDIA DGX SuperPOD, these new certifications provide NVIDIA Cloud Partners and enterprises with high-performance storage that meet all requirements for building state-of-the-art AI infrastructure. By leveraging Pure Storage, customers can confidently deploy AI factory infrastructure while ensuring performance and scalability.
The advent of large-scale AI and ML workloads has led to the rapid growth of GPU cloud providers that are scaling infrastructure to meet enterprise demand for accelerated computing. Enterprise AI projects rely on high-performance storage systems to keep up with the pace and scale of model training, inference, and retrieval-augmented generation (RAG) workflows. However, traditional storage solutions often limit GPU efficiency, creating bottlenecks in data access and directly impacting time to outcomes. In addition, these traditional storage architectures introduce complexities in managing storage at scale and struggle with multi-tenant environments, increasing operational inefficiencies and risk of security gaps.
With Pure Storage, organisations can deploy a storage solution that is compatible with NVIDIA accelerated computing and removes the barriers to real ROI and GPU workload efficiency. The Pure Storage platform eliminates bottlenecks and seamlessly handles multi-tenancy while promoting always-on quality of service (QoS). Pure Storage offers a solution with high-level read/write performance guarantees that can accelerate large-scale training and inference for AI workloads. Pure Storage also balances comprehensive and robust data storage with an easy-to-manage interface, further removing common complexities associated with legacy storage systems.
“GPUs have fast become a driving force behind the next wave of AI innovation. The incorporation of the NVIDIA AI Data Platform into FlashBlade provides the AI-ready storage necessary for peak performance. Additionally, our recent NVIDIA certifications affirm that Pure Storage is supporting the pace and scale that AI models need to create change. Through our purpose-built storage solutions, enterprise customers can easily and efficiently harness the power of AI to accelerate success,” said Rob Lee, Chief Technology Officer, Pure Storage.
“The new wave of AI reasoning will be powered by massive datasets and requires an always-on data processing pipeline. By integrating NVIDIA AI Data Platform capabilities into Pure Storage FlashBlade, enterprises can equip AI agents with near-real-time business data to achieve new levels of personalised customer service, operational efficiency, and unprecedented productivity,” said Rob Davis, Vice President, Storage Networking Technology at NVIDIA.
“For customers, a simpler path to deploying AI at scale means fewer complexities and faster innovation. That’s why Cisco partners with leaders like Pure Storage, which pushes AI data storage boundaries with offerings like FlashBlade//EXA and support for NVIDIA’s new reference design. We’re excited to extend this momentum through our FlashStack collaboration, integrating Cisco’s data centre infrastructure with Pure’s advanced storage to help businesses streamline AI workloads and unlock the full potential of their data,” said Jeremy Foster, GM and SVP for Cisco Compute.
Discussion about this post