GIGABYTE H263-S62 2U 4-Node Multi-Node Server

The GIGABYTE H263-S62 is a 2U 4-node multi-node server engineered for high-density, scale-out compute deployments in cloud, hyperscale, and enterprise data center environments. Each of the four independent nodes is built on the AMD EPYC 9004 series (Genoa) platform, supporting a single processor per node with up to 96 cores, delivering exceptional aggregate compute density within a compact 2U chassis footprint. The shared chassis architecture maximizes rack utilization while keeping per-node power and thermal management highly efficient through a centralized redundant fan and power subsystem design.

Each node accommodates DDR5 memory across multiple DIMM slots, supports NVMe and SATA storage options for flexible local storage configurations, and provides PCIe Gen 5 expansion for high-bandwidth networking or accelerator add-in cards. The H263-S62 features a tool-less node slide-out mechanism, enabling individual node servicing without disrupting neighboring nodes or requiring chassis removal from the rack. Out-of-band management is available per node through dedicated ASPEED AST2600 BMC controllers supporting IPMI 2.0, Redfish, and GIGABYTE's own server management utilities.

Targeted at hyperscale operators, colocation providers, and enterprise IT teams building high-density compute clusters, the H263-S62 is particularly well-suited for cloud-native workloads, web-scale application hosting, distributed databases, and HPC node-dense environments where maximizing compute per rack unit is a primary design objective. Its support for the AMD EPYC 9004 processor family brings PCIe 5.0 fabric, enhanced memory bandwidth, and improved core density that directly address the throughput and latency requirements of modern scale-out application architectures.

Ideal for

• Hyperscale and cloud service provider scale-out compute cluster deployment requiring maximum core density per rack unit
• High-density web hosting and application server consolidation in colocation and enterprise data centers
• Distributed database and in-memory caching workloads benefiting from high per-node core counts and DDR5 memory bandwidth
• High-performance computing (HPC) environments building tightly packed compute node pools for parallel processing tasks
• Content delivery and media streaming infrastructure requiring large numbers of independent compute nodes in minimal rack space
• Edge data center deployments and micro-data-center builds where physical footprint and power efficiency are critical constraints

Technical specifications

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