Latest news with #V-ScaleFlow
Forbes
2 days ago
- Business
- Forbes
Digital Storage Can Assist Or Foil AI
Data Center In this article we will look at some resent announcerments on digital storage and its use in AI training and inference. But first, an example of digital storage technology used to save humanity. Data Storage Saves the Day Digital archiving startup SPhotonix's 5D memory crystal was an important element in the plot of the latest Mission Impossible movie. The 360TB memory crystal was used to stop a rogue AI from destroying the world. In practice, SPhotonix stores data using a FemtoEtch nano-etching technology on a 5-inch glass substrate. Note that I am an advisor for SPhotonix. SPhotonix 5D Memory Cystal Digital storage technologies have been used in many movies and TV shows over the years, such as the StorageTek Tape library used in the 1994 Film 'Clear and Present Danger.' Hybrid AI Data Centers In practice data centers are generally using SSDs as primary storage in data centers, including for AI training applications. SSDs provide fast storage for refreshing data on the high bandwidth memory located close to the GPUs that directly support data processing. However, the cost for storing data on SSDs in data centers is about 6X higher than storing it on HDDs. This leads data centers to use HDDs for storing colder but useful data in a hierarchical storage environment. Data is moved back and forth from various storage technologies to optimize the balance of cost versus performance. Ultimately archived information in data centers that is not frequently used is kept on magnetic tape cartridges or optical storage. Vdura, formerly veteran storage company, Panasas, recently announced a white paper on digital storage for AI workloads and announced changes in their hybrid SSD and HDD storage offering to support HPC and AI workloads. The company is now offering QLC NAND flash SSDs combined with high-capacity HDDs with their global namespace parallel file system combined with object storage, offering multi-level erasure coding and fast key value storage. The image below shows the layout of this hybrid SSD and HDD storage system. Vdura Global Namespace Storage The Vdura Data Platform V11.2 includes a preview of V-ScaleFlow that enables data movement across QLC flash and high-capacity hard drives. This allows resource utilization, maximizes system throughput and provides efficient AI-scale workloads. In particular the company is using Phison Pascari 128TB QLC NVMe SSD with 30+TB HDDs to reduce flash capacity requirements by over 50% and lowing power consumption. Overall total cost of ownership is said to be reduced by up to 60%. The Vdura white paper goes into details on data storage and memory utilization in an AI application. The figure below shows an AI data pipeline which should have the storage system enable minimum GPU downtime. AI Data Pipeline The table below goes into detail on read, write, performance and data size requirements for various elements in an AI workload. These various elements can require from GBs to PBs of digital storage with various performance requirements. This favors a combination of storage technologies to support different elements in this workload. Element Characteristics in an AI Workflow The below image shows a sample storage node that can provide all-flash or hybrid SSD and HDD storage to support AI and HPC workloads with a global namespace and a common control and data plane. Vdura Storage Node Digital storage technology saved the world from a rogue AI in the latest Mission Impossible Movie. Combining SSDs and HDDs can enable modern AI workloads that optimize cost and performance.
Techday NZ
3 days ago
- Business
- Techday NZ
VDURA unveils HyperScaleFlow v11.2 to cut AI storage costs
VDURA has introduced Version 11.2 of its Data Platform, HyperScaleFlow, which brings enhancements to support artificial intelligence and high-performance computing workloads. The latest release delivers native Kubernetes Container Storage Interface (CSI) support, comprehensive end-to-end encryption, and the launch of VDURACare Premier, a support package that combines hardware, software, and maintenance under a single contract. A preview of the V-ScaleFlow capability is also included, which manages data movement between high-performance QLC flash and high-capacity hard drives to improve efficiency for AI-scale operations and reduce costs. According to VDURA, native CSI support eases multi-tenant Kubernetes deployments by enabling persistent-volume provisioning and management without scripting. The new end-to-end encryption feature provides security for data from transfer through to storage, including tenant-specific encryption per volume. VDURACare Premier offers comprehensive support through a single contract, covering hardware, software, and services such as a ten-year no-cost replacement policy for drives and 24-hour expert assistance. The V-ScaleFlow technology, currently in preview within the software, introduces an optimised data management layer. It dynamically orchestrates placement and movement of data between QLC SSDs, such as the Pascari 128TB, and high-density hard drives exceeding 30TB each. This approach aims to reduce flash capacity requirements by more than 50 percent and cut power consumption, which the company says delivers significant cost savings for organisations building AI data pipelines. The V-ScaleFlow system tackles industry challenges associated with write-intensive AI checkpoints and long-term data storage by using V-Burst to absorb demand spikes and write data sequentially to large NVMe drives, halving the amount of flash needed. For long-tail datasets and historic artefacts, the system moves data to high-capacity hard drives, which is intended to reduce both operational expenses and energy usage per petabyte stored. The VDURACare Premier bundle addresses the complexity seen in contracts that separate hardware, software, and maintenance through a combined package with risk-free coverage across a decade. Benefits highlighted for Version 11.2 and V-ScaleFlow include seamless data movement between flash and disks, optimised storage economics that can lower total cost of ownership by up to 60 percent, sub-millisecond latency for NVMe-class performance, and streamlined Kubernetes deployment for stateful AI workloads. Ken Claffey, Chief Executive Officer of VDURA, said: "V11.2 delivers the speed, cloud-native simplicity, and security our customers expect - while V-ScaleFlow applies hyperscaler design principles, leveraging the same commodity SSDs and HDDs to enable efficient scaling and breakthrough economics." VDURA stated that Data Platform V11.2 will become generally available on new V5000 systems during the third quarter of 2025. The full release of V-ScaleFlow is anticipated in the fourth quarter of 2025. Current V5000 users will have access to upgrade to Version 11.2 through an online update process. VDURA is presenting the capabilities of its data platform at ISC 2025, alongside partners such as Phison Pascari, Seagate Mozaic, Starfish, and Cornelis Networks. The company is hosting the VDURA AI Data Challenge, an event featuring strongman Hafþór Björnsson, which will allow attendees to engage with interactive data tasks and evaluate GPU-optimised data performance. Commenting on the technology, Michael Wu, President and General Manager of Phison U.S., said: "Phison has collaborated closely with VDURA to validate V-ScaleFlow technology, enabling seamless integration of our highest-capacity QLC Pascari enterprise SSDs in the VDURA Data Platform. V-Burst optimises write-intensive AI workloads, delivering exceptional performance and endurance while driving down costs - a game-changer for HPC and AI environments." Trey Layton, Vice President of Software and Product Management for Advanced Computing at Penguin Solutions, added: "Penguin Solutions is excited to see VDURA's V11.2 release and breakthrough features that include V-ScaleFlow, native CSI support, and end-to-end encryption that advance the operational goals of our enterprise, federal, and cloud customers. These enhancements simplify persistent storage orchestration across Kubernetes environments, ensure robust security without performance tradeoffs, and unlock compelling TCO improvements for organisations scaling AI and HPC workloads. VDURA continues to deliver a platform purpose-built for the future of real-time, inference-driven infrastructure."
