September 24, 2024 — Giga Computing, a subsidiary of GIGABYTE and an trade chief in generative AI servers and superior cooling applied sciences, right now introduced its first wave of GIGABYTE servers for Intel Xeon 6 Processors with P-cores. This new Intel Xeon platform is engineered to optimize per-core-performance for compute-intensive and AI intensive workloads, in addition to normal objective purposes. GIGABYTE servers for these workloads are constructed to attain the very best efficiency by superb tuning the server design to the chip design and to particular workloads.
All new GIGABYTE servers help Intel Xeon 6900-series processors with P-cores which have as much as 128 cores and as much as 96 PCIe Gen5 lanes. Moreover, for larger efficiency in reminiscence intensive workloads, the 6900-series expands to 12 channel reminiscence, and makes obtainable as much as 64 lanes CXL 2.0. Total, this modular SOC structure has nice potential with the flexibility to leverage a shared platform for operating each efficiency and effectivity optimized structure.
Basic Objective (Single-Socket) Servers
R164 Sequence: Compact 1U design that helps a single Intel Xeon 6 processor and a dual-slot GPU. Server variation inside is generally centered on the storage resolution with decisions of both 12x 2.5” bays or 4x 3.5”/2.5” bays for NVMe/SATA/SAS drives.
R264 Sequence: Increasing to a 2U chassis, this sequence additionally helps a single processor, however with the flexibility to help as much as 4x dual-slot Gen5 GPUs.
Basic Objective (Twin-Socket) Servers
R184 Sequence: Related storage decisions are potential just like the R164 sequence; nonetheless, with the R184 comes larger compute density unfold throughout two CPU sockets, and to accommodate for this there’s much less of a deal with the enlargement slots and eliminated dual-slot GPU help.
R284 Sequence: Shifting to a 2U chassis, the larger server top affords the flexibility to have each twin CPU sockets and help for dual-slot GPUs. The fashions inside the R284 sequence embody choices of as much as 24x 2.5” Gen5 drives or 12x 3.5”/2.5” combined storage.
Edge Servers
E264 Sequence: For these in want of quick chassis depth, this sequence retains the flexibility to have a 2U kind think about help of a single processor and as much as 4 dual-slot GPUs. The server depth has decreased by almost 200mm in comparison with general-purpose servers, but the general compute efficiency could be maintained with a tradeoff in much less storage bays.
E284 Sequence: With an emphasis on CPU compute, this sequence is a twin CPU socket configuration with 6x FHHL enlargement slots and 2x OCP NIC 3.0 slots for a wealth of I/O choices.
NVIDIA MGX Modularized Servers
XV24 Sequence: An NVIDIA OVX optimized server platform, this new twin socket sequence helps as much as 4 NVIDIA L40S GPUs and has extra FHFL slots for NICs and DPUs.
Excessive-density, Multi-node Servers
H374 Sequence: For the last word in dense CPU compute efficiency, this sequence helps as much as eight CPU sockets unfold throughout 4 nodes. Every node additionally has 4 low-profile expansions slots and both two or six 2.5” bays.
Server-grade Motherboards
MA34 Sequence: Excessive-quality server board filled with I/O choices, the MA34-CP0 is a board wealthy in each MCIO 8i and SlimSAS connectors, but it has 4 Gen5 enlargement slots and an OCP NIC 3.0 slot.
GIGABYTE continues to push the boundaries of server innovation, delivering options that meet the evolving calls for of contemporary computing. The brand new servers constructed on Intel Xeon 6900-series processors are designed to empower companies throughout varied sectors, offering unmatched efficiency, flexibility, and effectivity. As industries more and more depend on AI, cloud computing, and edge applied sciences, GIGABYTE’s dedication to aligning its server designs with the newest developments in processor know-how ensures that our prospects are well-equipped to deal with the challenges of tomorrow.
As extra is unpacked with the probabilities within the Intel Xeon 6 platform, GIGABYTE will prepared programs to match all workloads in addition to vertical markets. Moreover, within the coming months, GIGABYTE will announce a brand new GIGABYTE server that integrates Intel Gaudi 3 GPUs. This scalable and cost-effective resolution will goal AI workloads, together with generative AI inference workloads. And prospects can count on to see it at SC24 in Atlanta.
Uncover GIGABYTE options, go to https://www.gigabyte.com/Industry-Solutions/intel-xeon-6.
For queries or extra info, please GIGABYTE contact sales.
About Giga Computing
Giga Computing Expertise is an trade innovator and chief within the enterprise computing market. Having spun off from GIGABYTE, we preserve {hardware} experience in manufacturing and product design, whereas working as a standalone enterprise that may drive extra funding into core competencies. We provide a whole product portfolio that addresses all workloads from the information heart to edge together with conventional and rising workloads in HPC and AI to knowledge analytics, 5G/edge, cloud computing, and extra. Our longstanding partnerships with key know-how leaders be sure that our new merchandise would be the most superior and launch with new associate platforms. Our programs embody efficiency, safety, scalability, and sustainability.
Supply: GIGABYTE