Western Digital OpenFlex Data24 4000 NVMe-oF Storage Platform

The Western Digital OpenFlex Data24 4000 NVMe-oF Storage Platform (MPN: 1ES2392) is a 24-bay Ethernet Bunch of Flash (EBOF) appliance purpose-built for disaggregated, NVMe-over-Fabrics storage architectures. The system decouples flash storage resources from compute nodes, enabling independent scaling of storage capacity and performance across a fabric-attached infrastructure. Each drive bay accepts U.2 NVMe SSDs, and the platform exposes those drives natively over NVMe-oF using NVMe/TCP, allowing any connected host to address individual namespaces with near-local NVMe latency across a standard 100GbE Ethernet network.

The Data24 4000 integrates a self-contained storage services module that manages fabric connectivity, drive health, and namespace allocation without requiring a dedicated external storage controller or a host-side kernel driver beyond the standard Linux NVMe-oF initiator. The appliance ships with Western Digital's OpenFlex software stack, which provides RESTful API-driven provisioning, enabling automated orchestration through platforms such as Kubernetes and OpenStack. The 2U chassis accommodates up to 24 NVMe SSDs in hot-swap U.2 form factor bays and connects to the data network via dual 100GbE QSFP28 ports, delivering aggregate sequential throughput suited to the most demanding flash workloads.

Designed for hyperscale, cloud, and high-performance enterprise environments, the OpenFlex Data24 4000 is particularly well suited to workloads that benefit from storage disaggregation, including AI/ML training pipelines, high-frequency transactional databases, real-time analytics, and media rendering farms. The platform's shared-nothing, per-namespace access model allows multiple compute nodes to access distinct NVMe namespaces simultaneously without storage-layer contention, improving resource utilization and reducing stranded flash capacity compared to traditional direct-attached storage deployments.

Ideal for

• AI and machine learning training clusters requiring high-throughput, low-latency access to large training datasets from multiple GPU compute nodes simultaneously
• Disaggregated Kubernetes persistent storage environments where storage capacity must scale independently from compute node count
• High-frequency transactional database deployments (e.g., MySQL, PostgreSQL, MongoDB) demanding sub-100µs fabric-attached NVMe latency
• Real-time analytics platforms processing large in-flight datasets where bottlenecks at the storage layer would compromise query SLAs
• Media and entertainment post-production workflows requiring shared, high-bandwidth NVMe storage accessible across multiple editing workstations over Ethernet
• Private and hybrid cloud infrastructure consolidation projects replacing multiple direct-attached storage pools with a centrally managed, fabric-shared flash tier

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.