VDURA V5000 delivers AI-optimized all-flash storage with 1.5PB per U, zero-downtime scaling, and breakthrough metadata performance.
VDURA has launched its latest product, the V5000 All-Flash Appliance, to address the increasing demands of AI pipelines and the deployment of generative AI models. The V5000 appliance, driven by the VDURA V11 data platform, achieves high throughput that saturates GPU capabilities, ensuring strong data durability and availability while maintaining continuous operation year-round.
The VDURA V11 Software runs on the newly developed ‘F Node,’ a modular 1U platform featuring Intelligent Client-Side Erasure Coding. This platform offers more than 1.5PB of storage per rack unit and delivers scalability and reliability, making it ideal for handling the increasing storage demands of AI workloads.
VDURA’s technology provides simple scalability for AI environments. The V5000 All-Flash Appliance allows AI service providers to dynamically expand their infrastructure, unlike conventional AI storage solutions that rely on overprovisioning. Organizations can scale from a handful of nodes to thousands without downtime, enabling seamless growth as GPU clusters increase. The V5000 storage nodes integrate into existing setups, boosting both capacity and performance without disruption. By combining the V5000 All-Flash with the V5000 Hybrid, VDURA offers a unified, high-performance data infrastructure that supports every stage of the AI lifecycle—from rapid model training to long-term data retention.
V5000 Tech Highlights
- Three PCIe and one OCP Gen 5 slot for high-speed expansion connectivity.
- NVIDIA CX-7 SmartNICs for ultra-low latency data transfer.
- One AMD EPYC 9005 Series processor and 384GB of memory.
- Up to 12 U.2 128TB NVMe SSDs, providing more than 1.5PB per rack unit.
The VDURA Data Platform’s parallel file system (PFS) architecture, designed specifically for AI workloads at scale, overcomes the traditional bottlenecks associated with high-frequency checkpointing. VDURA’s Intelligent Client architecture ensures that AI storage maintains peak performance and reliability, employing lightweight client-side erasure coding without introducing unnecessary CPU load. This approach sets VDURA apart from alternative solutions that often impose heavy computational overhead.
The V5000 All-Flash Appliance delivers sustained performance across AI workloads, supported by advanced features such as:
- Seamless AI Storage Expansion: Easily scale storage as GPU clusters grow without downtime or overprovisioning.
- AI Checkpoint Optimization: Remove write bottlenecks that slow down AI training.
- Data Center Efficiency: Achieve over 1.5PB per U, reducing power, cooling, and footprint costs.
- RDMA Acceleration: Leverage GPU Direct and RDMA capabilities with more optimizations.
As Ken Claffey, CEO of VDURA, stated, “The V5000 All-Flash Appliance, powered by our next-gen V11 Data Platform, delivers not only the peak performance enterprises expect—but while high performance is necessary, it is not enough. AI workloads demand sustained high performance and unwavering reliability. That’s why we’ve engineered the V5000 to not just hit top speeds, but to sustain them—even in the face of hardware failures—ensuring every terabyte of data fuels innovation, not inefficiencies or downtime.”
By addressing the unique challenges of AI-scale data performance, VDURA sets new industry benchmarks. The combination of the V5000 All-Flash and V5000 Hybrid, backed by the advanced VDURA V11 Data Platform, provides an end-to-end solution that ensures uncompromising performance, scalability, and efficiency for AI workloads.
The VDURA V5000 All-Flash Appliance is now available for customer evaluation and qualification. Early deployments are taking place in key AI data centers. VDURA is collaborating with AI service providers to enhance performance and scalability. General availability is expected later this year.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
