1U rackmount
2 PCIe 5.0 x16 FHHL
Up to 12 hot-swap 2.5" NVMe/SAS/SATA drive bays
2 M.2 slots
Redundant up to 2000W Titanium Level M-CRPS
-1 RJ45 dedicated IPMI management port
- 2 USB 3.0 ports with USB 2.0 speed
- 1 USB 3.0 port with display port support
- ASPEED AST2600 BMC graphics
Single SP5 socket for AMD EPYC™ 9004 or 9005 Series processors, including those with AMD 3D V-Cache™ technology, up to 192 cores, up to 500W TDP2
12 DIMM slots; speeds up to 6000MT/s with up to 9TB
Regardless of your data center’s management approach, our open management APIs and tools are ready to support you. In addition to a dedicated IPMI port, and a Web IPMI interface per node, Supermicro® SuperCloud Composer software helps you configure, maintain, and monitor all of your systems using single-pane-of-glass management. If your DevOps teams prefer to use their own tools, industry-standard Redfish® APIs provide access to higher-level tools and scripting languages.
The A+ Server 2116GT-HNTF has all storage and I/O accessible from the front panel. Each node supports up to four U.2 NVMe or SATA drives and a front I/O card with options including dual 10 or 25 Gigabit Ethernet, or single 100 Gigabit Ethernet interfaces.
For even more demanding network needs, the left pair of drive bays can be interchanged for a Supermicro Advanced I/O Module (AIOM) cage that supports OCP 3.0 interfaces.
This Supermicro multi-node system is designed for applications that need a large number of discrete servers with high-speed interconnects for networked or clustered operations. They are ideal for virtualized and nonvirtualized applications including:
Hyperconverged infrastructure and scale-out storage applications where a balanced set of resources is key
High-performance computing including EDA simulation, computational fluid dynamics, and weather modeling
Content-delivery networks where a large number of network streams need to run in parallel
Back-end infrastructure for mobile devices including gaming, voice recognition, and mapping services
Cloud computing where a large number of cores are needed to deliver high performance to each virtual machines
Big data analytics that combine scale-out storage with the need for high compute capacity for data analysis
These 2U servers optimize compute, memory, and I/O resources to deliver maximum density—four single-socket nodes in only two rack units. In this front-I/O system, all storage, networking, and node trays are accessible from the cold aisle, simplifying installation and servicing in space-constrained environments.
Each of the four nodes host a single AMD EPYC 9005 or 9004 Series CPU with up to 192 cores, up to 16 DIMMs for a total of 6 TB of DDR5-6000 memory (with EPYC 9005 Series), up to four U.2 NVMe or SATA drives with PCIe 5.0 connectivity, two M.2 slots for boot drives, and a range of networking options to keep data flowing freely through these powerful servers. Dual redundant 2200W power supplies economize on power and cooling.
Maximize resource savings through shared power and cooling
AMD EPYC™ 9005 or 9004 Series processor with up to 6 TB of DDR5-6000 memory per node
Up to four 2.5" NVMe or SATA drives per node
PCIe 5.0 with CXL 2.0 support
Front I/O configuration enables field service from cold aisle to help reduce downtime
Flexible networking options with PCIe 5.0 OCP 3.0 interfaces
Redundant Titanium level shared power supplies
CDN
HPC
EDA
Cloud
Designed for maximum density, the GrandTwin™ server is built on a multi-node architecture for to exploit the performance of 4th and 5th Gen AMD EPYC processors for high single-CPU performance. The H14 update to the GrandTwin design includes the capability to support up to 160 cores in EPYC 9005 Series processors, higher thermal capability, and an optional PCIe 5.0 x16 AOIM slot. The GrandTwin system delivers high performance in a modular design that can be optimized for a wide range of options, with the capability to add or remove components as needed to match data center needs.