-
service Device Model
HN-Y6128V2
platform
Intel AI Computing service device standard system platform
processor
supports Intel LGA3647 platform processor, supports Intel Xeon 1st & 2nd gen dual processor
chipset
Intel C621/C622 series
Memory Standard
supports 16 Strip DDR4 2933/2666/2400/2133MHz spec server memory (depends on CPU), max support 2TB
graphics card
Aspeed AST2500 BMC Controller-1x DisplayPort 1.1a with max.resolution up to 1920x1200 @ 60Hz
Power Supply
can Select GRPS800W-3200W Power Supply, supports N+M Redundant
NIC
2 Intel 1210 Gigabit Ethernet port 1 x IPMI remote management port Realtek RTL8211E
Storage
supports 4 x 3.5 inch hard drive, supports 4/2.5 inch hard drive SSD SSD Drive, 3 x M.2 2280 NVME PCl Express 4.0x4
I/O port
2 x USB2.0 port, 2 x USB3.0 port, 2 x RJ45 LAN port, 1 x RJ45 management port, 1 x VGA Display Interface, 1 x COM port
RAID supports
supports RAID 0/1/5/10
work Environment
operating temperature:0°C~55°℃(32℉~131°F) work Humidity:10%-95% @40°C Non-Condensing
Chassis parameters
Color: Black, Material: SGCC Galvanized Steel, cooling Thermal: 3 x Active Fan cooling Thermal, Complete System dimensions: 850L*437.6W*176H
Ordering Information
standard
standard system platform: system Chassis +3*12038 Thermal header Removable Fan +5 Power Backplane +11 PCIE 4.0GPU baseboard +CPU cooler + Stress Tester
can Select configuration
CPU: Intel LGA3647 platform processor, supports Intel Xeon Gen 1/2 Generation dual processor
memory: ECC DDR4/ server memory can Select
graphics card: multiple types single-width full high full-length mainstream A1 graphics card card can Select
hard drive: max support 14 x SATA 3.0, among which 12 x SATA requires SAS SFF8087 RPM SATA adapter cable RPM out
OEM service
provide solution customized service
-
User Manual
Installation Video
-

HN-Y6128V2
· balanced architecture, outstanding performance: max support 8 single-width full-height full-length GPU, significantly boosts heterogeneous computing
· rich ecosystem, system optimization: supports various mainstream AI accelerator cards and leading DL frameworks flexibly adapt to business scenarios
· flexible configuration, easy maintenance: modular redundant design, most components tool-free serviceable, quick disassembly reduces maintenance time, specifically as heavy-load graphics processing tasks design, configuration high-performance GPU, specifically as computing intensive applications, such as deep learning, 3D rendering, scientific simulation, video editing, etc.
· compute density: internal support for multiple GPU, as needs scenarios provides high-density computing with and expansion capability
· multiple connection methods: supports direct connection /switch architecture, based on different industry use scenarios, provides more suitable VRAM access methods, reduces latency and bandwidth bottleneck
· high bandwidth, low latency: expandable with multiple network adapter interfaces (10GbE, 25GbE, 40GbE, 100GbE, 200GbE, 400GbE) with supports high-bandwidth, low-latency data transfer
· application areas: widely used at people work intelligent, deep learning, Smart Education, Smart City, Smart Healthcare, high-performance computing, data analysis, Metaverse etc. should using scenarios