AMD Instinct MI355X

GIGAPOD AI AC POD Scale Solution
  • Built on the AMD Instinct MI355X platform
  • 32 x GIGABYTE G893 8U GPU servers
  • 256 x AMD Instinct MI355X OAM GPUs
  • 64 x AMD EPYC 9005/9004 Server Processors
  • 2.3 TB HBM3E GPU memory per server
  • Up to 70 kW rack power consumption
  • Reserved space for InfiniBand / Ethernet switches
  • Compatible with GIGABYTE POD Manager software for cluster management / orchestration
  • Supports flexible storage with GIGABYTE storage servers and validated partner solutions
儲存
此產品已儲存至最愛清單
我的收藏產品
詢價清單

Configuration

PAA3 Series
Cooling TypeAir Cooling
Racks per Scalable Unit
(Compute + Management)
8+1
Rack Height42U / 48U
Server Model
Servers per Rack32 x 8U GPU servers
CPUAMD EPYC 9005 Server Processors
AMD EPYC 9004 Server Processors

Total 64 processors
GPUAMD Instinct MI355X
- Total 256 x AMD Instinct MI355X OAM
SwitchFlexible validated InfiniBand / Ethernet switch configurations
PDU4 x PDUs
PDU SpecificationsInput: 415 V, 63 A, 3-phase
Power Consumption per RackUp to 70 kW
ManagementOptional GIGABYTE POD Management (GPM) software suite:
- Essential single-node BMC for real-time hardware health and remote system oversight
- Unified cluster-scale management and deep telemetry for high-density deployments
- Total POD-level orchestration for infrastructure provisioning and environmental monitoring
Storage SolutionsFlexible validated partner solutions with GIGABYTE storage servers
 
Cooling Type
Racks per Scalable Unit
(Compute + Management)
Rack Height
Server Model
Servers per Rack
CPU
GPU
Switch
PDU
PDU Specifications
Power Consumption per Rack
Management
Storage Solutions
PAA3 Series
Air Cooling
8+1
42U / 48U
32 x 8U GPU servers
AMD EPYC 9005 Server Processors
AMD EPYC 9004 Server Processors

Total 64 processors
AMD Instinct MI355X
- Total 256 x AMD Instinct MI355X OAM
Flexible validated InfiniBand / Ethernet switch configurations
4 x PDUs
Input: 415 V, 63 A, 3-phase
Up to 70 kW
Optional GIGABYTE POD Management (GPM) software suite:
- Essential single-node BMC for real-time hardware health and remote system oversight
- Unified cluster-scale management and deep telemetry for high-density deployments
- Total POD-level orchestration for infrastructure provisioning and environmental monitoring
Flexible validated partner solutions with GIGABYTE storage servers

[#1] 本網站所提供之產品規格、圖片及其他資訊僅供參考,如有差異應以實際產品為準。技嘉科技保留隨時修改內容的權利,並不對因使用上述資料而導致的任何損失承擔責任。

[#2] 本產品所標示之效能數據為晶片廠商或相關介面標準制定組織所公佈的最大理論值,實際效能可能因系統配置及使用環境而有所不同。

[#3] 本文所使用的所有商標及企業識別標誌均為其合法所有者的財產。