VDURA, a leader in data infrastructure for AI and HPC, today has announced a groundbreaking release of its VDURA Data Platform software, designed to set new industry benchmarks in performance, scalability, and data management simplicity. This robust software solution addresses the surging demands of AI and high-performance computing, empowering enterprises and service providers to efficiently manage data, achieve high-speed performance, and accelerate AI initiatives without the traditional trade-offs in reliability and scalability.
The latest VDURA Data Platform is a significant modernization on the previous releases with a move to a fully parallel, microservices-based architecture, a new flash-optimized metadata engine and an enhanced object storage layer. Together, these features ensure unmatched efficiency, scalability, and durability across on-premises, cloud, and hybrid environments.
Software-Driven Infrastructure for Modern AI and HPC
“As AI workloads expand, data infrastructure must support massive write throughput for frequent GPU checkpointing, rapid random reads for RAG-driven inference, billions of files of varying sizes, and substantial capacity for large training datasets—all while ensuring absolute data protection,” said Ken Claffey, CEO of VDURA. “VDURA’s platform is purpose-built to address these demands with >11 nines of durability, exceptional performance, and uncompromising availability. Our solution is not only powerful and scalable but also easy to manage and economically efficient, giving organizations confidence that their valuable insights are secure and always accessible.”
Key Software Innovations
- VeLO: Velocity Layered Operations – VeLO is a new Key Value Store used within the platforms Director layer for handling small files and metadata operations. It supports the diverse range of workloads found in AI. It is optimized for flash and delivers up to two million IOPS per instance. It supports an infinitely scalable number of VeLO instances in the same Global Namespace.
- VPOD: Virtualized Protected Object Device – VPODs are discrete, virtualized, and protected storage units, forming the foundation of data storage in hybrid nodes. These infinitely scalable VPOD instances operate within a unified Global Namespace, providing high performance and flexible scalability. Data is safeguarded through erasure coding across multiple VPODs in the cluster, with an optional, additional layer of erasure coding within each VPOD for enhanced protection. This multi-layered approach achieves up to 11 nines of durability with minimal overhead. Data reduction services further optimize efficiency, reducing costs and total ownership expenses.
- Advanced Data Management and Placement – The platform is now deployed as discrete microservices, simplifying its seamless deployment across thousands of nodes, helping ensure linear performance scalability with an infinitely expandable global namespace. The architecture also introduces intelligent data orchestration capabilities that enable optimal data placement across discrete tiers of storage but all within a single unified data and control plane within a single, unified namespace.
Certified Hardware Platforms for Optimized Performance and Flexibility
To complement its powerful software, VDURA also introduces the V5000 Certified Platform, designed to unlock the full potential of the VDURA Data Platform. This hardware platform combines modular storage nodes that can be configured for maximum throughput, capacity, or IOPS, depending on the user’s requirements.
Highlights of the V5000 Platform:
- 1U Director Nodes powered by the latest AMD EPYC 9005 processors, NVIDIA ConnectX-7, Broadcom 200Gb Ethernet and SAS 4 adapters and Phison PCIe NVMe SSDs, optimized for high IOPS and metadata operations through VeLO instances.
- Hybrid Storage Nodes incorporating the same 1U server used with the Director Node and 4U JBODs running VPOD instances for cost-effective bulk storage with high performance and reliability.
Industry-Leading Composability for Any Budget
VDURA’s platform supports a flexible ratio of HDD to Flash device content, allowing enterprises to optimize highly composable configurations for specific cost, performance, and durability needs. This flexibility empowers users to balance capacity and throughput while maintaining a consistent global namespace, making the VDURA Data Platform suitable for diverse capacity, price and performance requirements.
Future-Proofed for Continued Innovation
As part of VDURA’s dedication to continuous improvement, the Data Platform is set to expand further with upcoming releases. In 2025, VDURA will introduce the Cloud Edition, enabling seamless cloud and hybrid deployments. Additionally, a new All-Flash platform, designed for ultra-high IOPS, will be available by the second half of 2025, catering to the needs of AI and HPC environments that demand the highest data processing speeds.
VDURA will be at SC24 in Atlanta Georgia from November 17 – 22 at Booth #2209. For more information on the vision and technology behind VDURA, please stop by our booth or visit www.vdura.com and LinkedIn.
Supporting Quotes
Proven Success Across Industries
“After a thorough evaluation, we selected VDURA’s Data Platform and V5000 for its superior price/performance and durability. As a long-time BeeGFS user, we understand the challenges of managing high-performance storage at scale,” said Prof. Volker Lindenstruth, Goethe University, Frankfurt. “With our new AMD GPU cluster, we needed exceptional storage performance across diverse workloads, without sacrificing ease of use, reliability, or budget constraints—which all-flash solutions couldn’t meet. VDURA’s hybrid architecture delivered the ideal balance of price/performance and low management overhead, making it the perfect choice.”
— Prof. Volker Lindenstruth, Goethe University, Frankfurt
Excellence in Partner Collaboration
“Penguin Solutions is excited about VDURA’s new high-performance storage solution. With its microservices-based portability, this new platform aligns well to meet the increased performance, and scalability demands of today’s AI and HPC workloads,” said Dara Ambrose, vice president of engineering at Penguin Solutions. “We’re confident when deploying with VDURA’s VeLO and the V5000 that it will meet even the most demanding data challenges, empowering users to push the limits of innovation.”
— Dara Ambrose, Vice President of Engineering, Penguin Solutions
Industry Analyst
“While modern HPC and AI workloads are driving significant performance requirements, much more is demanded from storage solutions to keep computing resources, both CPUs and GPUs, fully utilized. Reliable storage that can provide data when it’s needed with complete confidence in its integrity is just as important, if not more so, than absolute performance,” observes Mark Nossokoff, Research Director and Lead Storage Analyst at Hyperion Research. “VDURA’s focus on coupling enterprise-class storage solution durability and availability with GPC and AI performance capabilities seeks to address the increasingly demanding requirements of these modern workloads.”
— Mark Nossokoff, Research Director and Lead Storage Analyst, Hyperion Research
About VDURA
VDURA is at the forefront of AI and HPC data storage and management, catering to on-premises, public cloud, and hybrid environments. Renowned for its unparalleled blend of performance, durability, and reliability, VDURA’s Data Platform builds upon a legacy of pioneering work in unlocking the full potential of data.
VDURA recently launched the Velocity Partner Program, empowering partners with comprehensive tools, resources, and support to accelerate growth and drive customer innovation. By joining the program, partners gain access to exclusive training, marketing materials, and joint go-to-market opportunities, ensuring they can deliver the highest value solutions to their clients. To learn more please visit VDURA.com.
View source version on businesswire.com: https://www.businesswire.com/news/home/20241112069177/en/