Quantum Announces Support for NVIDIA GPUDirect Storage with Myriad All-Flash File System
Quantum (NASDAQ: QMCO) has announced the development of a new parallel file system client for their Myriad all-flash file system, featuring NVIDIA GPUDirect Storage capability. The innovative solution enables on-the-fly client node deployments with cross-platform compatibility, including NVIDIA Grace Hopper architectures.
The new client installs directly on GPU-equipped servers, creating operational Myriad GPU Nodes that establish direct RDMA data paths between storage and GPU memory. This bypasses CPU bottlenecks and supports multiple connection options including SMB, NFS, and planned S3 access.
The system will be available for evaluation through Quantum's Early Access Program in Q1 2025, with general availability planned for the second half of 2025. The solution targets AI/ML, HPC visualization, and video rendering applications.
Quantum (NASDAQ: QMCO) ha annunciato lo sviluppo di un nuovo client per sistemi di file paralleli per il loro file system Myriad completamente flash, dotato della tecnologia NVIDIA GPUDirect Storage. La soluzione innovativa consente implementazioni di nodi client al volo con compatibilità tra piattaforme, inclusa l'architettura NVIDIA Grace Hopper.
Il nuovo client si installa direttamente sui server dotati di GPU, creando nodi GPU Myriad operativi che stabiliscono percorsi di dati RDMA diretti tra lo storage e la memoria GPU. Questo evita i colli di bottiglia della CPU e supporta diverse opzioni di connessione tra cui SMB, NFS e l'accesso S3 pianificato.
Il sistema sarà disponibile per la valutazione attraverso il Programma di Accesso Anticipato di Quantum nel Q1 2025, con disponibilità generale prevista per la seconda metà del 2025. La soluzione è destinata ad applicazioni di AI/ML, visualizzazione HPC e rendering video.
Quantum (NASDAQ: QMCO) ha anunciado el desarrollo de un nuevo cliente de sistema de archivos paralelo para su sistema de archivos Myriad totalmente flash, que cuenta con la capacidad de NVIDIA GPUDirect Storage. La solución innovadora permite implementaciones de nodos cliente al instante con compatibilidad entre plataformas, incluyendo arquitecturas NVIDIA Grace Hopper.
El nuevo cliente se instala directamente en servidores equipados con GPU, creando nodos GPU Myriad operativos que establecen rutas de datos RDMA directas entre el almacenamiento y la memoria de la GPU. Esto evita cuellos de botella en la CPU y soporta múltiples opciones de conexión, incluyendo SMB, NFS y acceso S3 planificado.
El sistema estará disponible para evaluación a través del Programa de Acceso Anticipado de Quantum en Q1 2025, con disponibilidad general prevista para la segunda mitad de 2025. La solución está dirigida a aplicaciones de IA/ML, visualización HPC y renderización de video.
Quantum (NASDAQ: QMCO)는 Myriad 올플래시 파일 시스템을 위한 새로운 병렬 파일 시스템 클라이언트 개발을 발표했으며, 이는 NVIDIA GPUDirect Storage 기능을 지원합니다. 이 혁신적인 솔루션은 NVIDIA Grace Hopper 아키텍처를 포함한 크로스 플랫폼 호환성을 갖춘 클라이언트 노드를 즉시 배포할 수 있습니다.
새로운 클라이언트는 GPU 장착 서버에 직접 설치되어, 스토리지와 GPU 메모리 간에 직접 RDMA 데이터 경로를 설정하는 운영 Myriad GPU 노드를 생성합니다. 이렇게 하면 CPU 병목 현상을 피할 수 있으며, SMB, NFS 및 계획된 S3 액세스를 포함한 여러 연결 옵션을 지원합니다.
이 시스템은 2025년 1분기(Q1 2025)에 Quantum의 조기 액세스 프로그램을 통해 평가 가능하며, 2025년 하반기에 일반 가용성이 계획되어 있습니다. 이 솔루션은 AI/ML, HPC 시각화 및 비디오 렌더링 애플리케이션을 목표로 합니다.
Quantum (NASDAQ: QMCO) a annoncé le développement d'un nouveau client de système de fichiers parallèle pour leur système de fichiers Myriad entièrement flash, doté de la capacité NVIDIA GPUDirect Storage. Cette solution innovante permet des déploiements de nœuds clients à la volée avec compatibilité multiplateforme, y compris les architectures NVIDIA Grace Hopper.
Le nouveau client s'installe directement sur des serveurs équipés de GPU, créant des nœuds GPU Myriad opérationnels qui établissent des chemins de données RDMA directs entre le stockage et la mémoire GPU. Cela contourne les goulets d'étranglement de CPU et prend en charge plusieurs options de connexion, y compris SMB, NFS et un accès S3 prévu.
Le système sera disponible pour évaluation via le Programme d'Accès Anticipé de Quantum au Q1 2025, avec une disponibilité générale prévue pour le second semestre 2025. La solution vise des applications d'IA/ML, de visualisation HPC et de rendu vidéo.
Quantum (NASDAQ: QMCO) hat die Entwicklung eines neuen Parallel-Dateisystem-Clients für ihr Myriad-All-Flash-Dateisystem angekündigt, das über die NVIDIA GPUDirect Storage Fähigkeit verfügt. Die innovative Lösung ermöglicht die sofortige Bereitstellung von Client-Knoten mit plattformübergreifender Kompatibilität, einschließlich der NVIDIA Grace Hopper-Architekturen.
Der neue Client installiert sich direkt auf GPU-ausgestatteten Servern und erstellt Betreibende Myriad-GPU-Knoten, die direkte RDMA-Datenpfade zwischen Speicher und GPU-Speicher herstellen. Dadurch werden CPU-Flaschenhälse umgangen und es werden mehrere Verbindungsoptionen unterstützt, darunter SMB, NFS und geplante S3-Zugriffe.
Das System wird im Q1 2025 über Quantums Early Access Program zur Verfügung stehen, mit einer allgemeinen Verfügbarkeit, die für die zweite Hälfte von 2025 geplant ist. Die Lösung richtet sich an AI/ML, HPC-Visualisierung und Video-Rendering-Anwendungen.
- Development of new technology enabling direct GPU-storage communication, potentially improving performance
- Cross-platform compatibility expansion including NVIDIA Grace Hopper architecture
- Flexible deployment options with multiple connection choices (SMB, NFS, planned S3)
- Early Access Program launch indicating market readiness and customer validation
- Product won't be generally available until second half of 2025, indicating extended time to market
- Early stage development with unproven market performance
Insights
Quantum's new parallel file system client for Myriad represents a significant technological advancement in AI/ML infrastructure. The integration with NVIDIA GPUDirect Storage and support for Grace Hopper architectures positions Quantum to capture growing demand in the AI storage market. The direct RDMA data path between storage and GPU memory, bypassing CPU bottlenecks, could deliver substantial performance improvements for AI workloads.
The innovative approach of transforming host workstations into operational Myriad GPU Nodes offers flexibility and scalability advantages over traditional parallel file system implementations. However, with general availability not expected until H2 2025, competitors have time to develop similar capabilities. The early access program starting Q1 2025 will be important for validating real-world performance benefits.
This development aligns with the rapidly growing AI infrastructure market, where storage solutions optimized for GPU workloads are in high demand. Quantum's market cap of
The flexible deployment model could reduce infrastructure costs and complexity for customers, potentially expanding Quantum's addressable market. However, the extended timeline to general availability poses some risk, as the fast-moving AI market could see new competitive solutions emerge before Myriad's full release.
Highly Parallel Direct Client Unlocks New Potential for AI/ML, HPC, and Life Sciences Customers
(Graphic: Business Wire)
The Myriad parallel client design takes a groundbreaking approach to building out AI/ML infrastructure by installing directly on customer servers or workstations equipped with high-performance GPU cards, transforming their host workstation into operational Myriad GPU Nodes. By leveraging NVIDIA’s Magnum I/O™ GPUDirect Storage technology, the client establishes a direct RDMA data path between storage and GPU memory, bypassing CPU bottlenecks to enable exceptional performance.
Multiple client nodes can be added to a Myriad cluster at any time, and are a powerful new connection option alongside server message block (SMB), network file system (NFS), and planned S3 access that can be added to any Myriad share point on-demand, giving customers the widest range of connection choice and flexibility available for all-flash file systems.
"Myriad’s parallel client development is guided by our vision to make Myriad the most capable, most flexible and easiest to use all-flash storage solution,” said Jeff Mulder, chief development officer, Quantum. “The new client brings unique capabilities by running as a fully integrated GPUDirect Node on client systems. Unlike traditional parallel file system clients, which rely on other appliances to perform these operations on their behalf, Myriad’s client performs its own metadata operations, data reduction, and data protection operations. This approach minimizes common bottlenecks, lets customers maximize their GPU investment across architectures, and allows performance to scale with the number of parallel clients accessing the Myriad system.”
The new Myriad parallel file system client will be made available for evaluation through Quantum’s Early Access Program. This program is an integral part of Quantum’s product development process, enabling influential customers to test and validate new features and provide real-world workflow feedback that will benefit all users.
“The growing demands of massive data analysis and AI/ML pipelines are driving the need for more agile, on-demand infrastructure solutions that are simple to implement and also ready for advanced technologies like NVIDIA GPUDirect Storage,” said Guy Currier, CTO, Futurum Research. “Organizations are increasingly looking for scalable, high-performance systems that can adapt to evolving workflows without requiring extensive new deployments. Simplicity, ease of use, and future-proof adaptability are critical factors in addressing these needs effectively.”
Quantum’s wide network of system integrators is ready to help integrate Myriad’s capabilities for mission critical content production needs such as animation/visual effects (VFX) rendering and AI/ML actions for rich media and video at scale.
"We are extremely impressed with the performance and capabilities of Myriad after testing it in our labs,” said Lance Hukill, chief commercial officer for CHESA, a leading integrator for media and entertainment customers. “The innovative approach to a parallel file system client will help our customers tackle the most demanding workloads that cannot be addressed with traditional legacy NAS storage systems and gives us tremendous flexibility in building customer solutions. We especially look forward to testing the new capabilities for large-scale high-resolution video rendering and development of new AI/ML tools to extract insights from large content lakes.”
Existing Myriad customers are eligible to participate in the Early Access Program for the new parallel file system client with NVIDIA GPUDirect Storage support upon its release in Q1 2025, with general availability in the second half of 2025. New prospects interested in Myriad and participating in the Early Access Program can contact Quantum at myriad-client-ea@quantum.com. For more information about Myriad, visit www.quantum.com/myriad.
About Quantum
Quantum delivers end-to-end data management solutions designed for the AI era. With over four decades of experience, our data platform has allowed customers to extract the maximum value from their unique, unstructured data. From high-performance ingest that powers AI applications and demanding data-intensive workloads, to massive, durable data lakes to fuel AI models, Quantum delivers the most comprehensive and cost-efficient solutions. Leading organizations in life sciences, government, media and entertainment, research, and industrial technology trust Quantum with their most valuable asset – their data. Quantum is listed on Nasdaq (QMCO). For more information visit www.quantum.com.
Quantum and the Quantum logo are registered trademarks of Quantum Corporation and its affiliates in
Forward-Looking Information
The information provided in this press release may include forward-looking statements within the meaning of Section 27A of the Securities Act of 1933 and Section 21E of the Securities Exchange Act of 1934. These forward-looking statements are largely based on our current expectations and projections about future events affecting our business. Such forward-looking statements include, in particular: Quantum Myriad and our plans, objectives and intentions that are not historical facts generally. These forward-looking statements may be identified by the use of terms and phrases such as “anticipates”, “believes”, “can”, “could”, “estimates”, “expects”, “forecasts”, “intends”, “may”, “plans”, “projects”, “targets”, “will”, and similar expressions or variations of these terms and similar phrases.
Additionally, statements concerning future matters and other statements regarding matters that are not historical are forward-looking statements. Investors are cautioned that these forward-looking statements relate to future events or our future performance and are subject to business, economic, and other risks and uncertainties, both known and unknown, that may cause actual results, levels of activity, performance or achievements to be materially different from those expressed or implied by any forward-looking statements.
These forward-looking statements involve risks and uncertainties that could cause actual results to differ materially from those projected, including without limitation, the following: Quantum Myriad and the impact of these factors on our performance and outlook. See also other risks that are described in “Risk Factors” in our filings with the Securities and Exchange Commission (the SEC), including its Annual Report on Form 10-K filed with the SEC for the fiscal year ended March 31, 2024, and any subsequent reports filed with the SEC. We do not intend to update or alter these forward-looking statements, whether as a result of new information, future events or otherwise, except as required by applicable law or regulation.
View source version on businesswire.com: https://www.businesswire.com/news/home/20241218337090/en/
Sara Beth Fahey
Matter Communications
quantum@matternow.com
401.351.9507
Source: Quantum Corporation
FAQ
When will Quantum's (QMCO) new Myriad parallel file system client be generally available?
What are the key features of QMCO's new Myriad parallel file system client?
How does QMCO's new Myriad client improve GPU performance?
What industries and applications is QMCO's new Myriad client targeting?
How can customers participate in QMCO's Myriad Early Access Program?
