serverbastion.com
● BRAND-NEW · FUSIONSERVER G5500 V6
xFusion FusionServer G5500 V6 — proven 4U AI server with up to 8 FHFL dual-width GPUs
The FusionServer G5500 V6 is the previous-generation 4U accelerated server in xFusion’s flagship line, powered by two 3rd-Generation Intel Xeon Scalable (Ice Lake) processors with 32 DDR4 DIMMs up to 3200 MT/s (up to 8 TB) and 8 full-height full-length dual-width GPU cards. It keeps the rugged X-lattice chassis, one-click topology switchover and GPUDirect support, and adds up to four 2000–3000 W PSUs with 6–8 redundant fans. A brand-new G5500 V6 gives you proven eight-card density and enterprise availability for AI training, inference, video analysis and HPC at a lower platform cost than the V7.
Proven accelerator density for cost-conscious AI projects
The G5500 V6 keeps the 4U eight-card form factor and redundant design on a mature, well-supported Ice Lake platform.
AI training
Up to eight FHFL dual-width accelerators, including NVIDIA A100-class cards, for deep-learning training on a value platform.
GPU clusters
Up to three OCP 3.0 NICs and a switchable topology let cost-optimized nodes form a sizeable accelerator pool.
Video & VDI
Dense professional GPUs support large virtual desktop, rendering and video-analysis deployments without flagship-platform pricing.
HPC & databases
Scientific computing, simulation and accelerated databases run on 3rd-Gen Xeon plus up to eight cards.
G5500 V6 technical specifications
Previous-generation 4U accelerator platform from the official datasheet; configured to your CPU, GPU, memory and storage needs.
| Specification | xFusion FusionServer G5500 V6 |
|---|---|
| Form factor | 4U AI server, dual-socket |
| Processors | 2 × 3rd-Generation Intel Xeon Scalable (Ice Lake), up to 270 W TDP per processor |
| Chipset | Intel C621A |
| Memory | 32 × DDR4 DIMM slots, up to 3200 MT/s, up to 8 TB total |
| GPU cards | 8 × FHFL dual-width GPU cards (also supports FHFL/HHHL single-width cards) |
| PCIe expansion | Up to 16 × slots: 12 × PCIe 4.0 x16, 3 × OCP, 1 × RAID |
| Network | 3 × OCP 3.0 NICs, configurable with orderly hot-swap |
| Local storage | 24 × 2.5/3.5" drives (maximum 4 × NVMe); dual M.2 SSDs |
| RAID | RAID 0/1/10/5/50/6/60; supercapacitor cache protection; level migration and drive roaming |
| Cooling | 6 or 8 × hot-swap counter-rotating custom fans in N+1 redundancy |
| Power | 4 × 2000 W Platinum/Titanium, 2600 W Platinum or 3000 W Titanium PSUs in N+N/N+M redundancy |
| Management | iBMC: Redfish, SNMP, IPMI 2.0, HTML5/VNC KVM; optional FusionDirector |
| Security | Power-on/admin passwords, TPM 2.0, security bezel, secure boot, chassis intrusion |
| Operating temperature | 5°C to 35°C, compliant with ASHRAE Class A1/A2 |
| Certification | CE, UL, CCC, FCC, VCCI and RoHS |
| Operating systems | Windows Server, SLES, RHEL, Ubuntu, openEuler and other supported systems |
| Dimensions (H×W×D) | 175 × 447 × 898 mm (6.89 × 17.60 × 35.35 in.) |
Eight-card performance on a mature, efficient platform
The G5500 V6 balances well-proven Xeon Scalable host compute, PCIe 4.0 expansion and dense accelerator capacity with reliable cooling.
Ice Lake
Two 3rd-Gen Xeon Scalable CPUs at up to 270 W provide strong core counts and PCIe 4.0 lanes for eight cards.
Up to 8 FHFL
Eight full-height dual-width accelerators in a 4U chassis give high AI and graphics density per node.
Great value
A mature platform with A100-class support delivers much of the capability at a lower entry cost.
Proven cooling
6–8 counter-rotating fans and up to four high-efficiency PSUs keep dense builds stable over long runs.
The same rugged 4U chassis on a proven platform
Modular risers, hot-swap fans and redundant power carry over, with Ice Lake, DDR4 and PCIe 4.0 host connectivity.
Internally the G5500 V6 follows the same proven layout as the V7: front hot-swap storage, a wall of 6 or 8 counter-rotating fans, and rear risers carrying up to eight FHFL dual-width accelerators. The motherboard uses 3rd-Generation Xeon Scalable sockets with 32 DDR4 slots, and up to three OCP NICs preserve standard PCIe slots. A one-click switch changes the topology between cascaded and balanced layouts, and fans and power supplies are hot-swap and redundant, giving the platform enterprise availability while its mature components keep build cost attractive.
Typical G5500 V6 configurations
Reference builds using supported, proven parts; we adjust GPU count, topology and power to the workload and verify the configuration before shipment.
Value eight-card training node
- 2 × Xeon Platinum 8380 (Ice Lake)
- 512 GB DDR4 across 16 DIMMs
- 8 × NVIDIA A100/A800 80 GB dual-width
- 4 × NVMe SSD + dual M.2 boot
- 3 × OCP 100 GbE · 4 × 3000 W Titanium
Dense VDI and rendering node
- 2 × Xeon Gold 6342
- 512 GB DDR4 across 16 DIMMs
- 6 × NVIDIA A40-class professional GPUs
- 8 × 3.5" SATA drives + dual M.2
- 2 × OCP 25 GbE · 4 × 2600 W Platinum
Cost-effective inference node
- 2 × Xeon Gold 6338
- 256 GB DDR4 across 8 DIMMs
- 4 × NVIDIA A30 / T4-class accelerators
- 4 × NVMe SSD + dual M.2 boot
- 2 × OCP 25 GbE · 4 × 2000 W PSU
Deployment, power and delivery for value GPU nodes
Cost-effective GPU clusters still need sound rack, power and logistics planning; we provide the same deployment support for V6 builds as for the newer platform.
Rack and airflow planning
We confirm rack depth for the 898 mm chassis, provide guide rails and plan standard front-to-back containment for 5–35°C operation, so V6 nodes install in conventional racks without liquid cooling.
Power sizing
With PSU options up to four 3000 W Titanium units, we specify the configuration for the selected eight-card build and advise on per-rack power, PDU and circuit capacity under load.
Export delivery
We return configured quotes within one working day, align firmware across batches, and arrange export packaging and documentation so brand-new V6 nodes arrive ready to rack and integrate.
How to choose the right G5500 V6 configuration
Eight-card value training
Use the full FHFL dual-width layout with A100-class cards and maximum memory when training on a budget.
Dense graphics and VDI
Use professional GPUs sized to concurrent users, with redundant storage for profiles, templates and rendered output.
Inference and analytics
Use fewer accelerators with OCP networking and NVMe for cost-effective video analysis and serving.
Every G5500 V6 is checked before it ships
Independent quality assurance signs the build only after every stage passes. The same checklist runs on every order whether it is one unit or a batch deployment, so the systems arriving in your data centre match the configuration and performance you were quoted.
Configuration verification
We check every CPU, GPU, memory module, drive, RAID controller and network adapter against the BOM and the agreed firmware levels before assembly.
Assembly and cable management
Components are seated and torqued to spec, power and signal cables routed for the front-to-back airflow path, and GPU retention and power cabling double-checked.
Full POST and firmware checks
The complete configuration boots through POST; CPU, memory and all accelerators are enumerated, and BIOS/BMC and RAID firmware versions are confirmed.
Extended burn-in under load
The server runs a high-utilization compute and GPU stress test to surface thermal, power or memory faults long before it reaches your rack.
Storage and RAID validation
RAID arrays are built and verified, hot-swap behaviour tested, and boot volumes confirmed independent of data drives.
Final independent QC sign-off
A second engineer reviews the test record and configuration, then signs off and seals the unit for export packaging and shipment.
Tested, documented, warranted. A test record accompanies every build and the 3-year warranty starts only after the unit passes final QC. Our team supports remote diagnosis and replacement parts through the warranty terms.
Backed by warranty, tested and export-ready
Warranty on a new server
Every G5500 V6 we ship is a brand-new system covered by a 3-year warranty, configured and burn-in tested before dispatch.
FHFL dual-width GPUs
Eight full-height dual-width accelerators on a proven Ice Lake / PCIe 4.0 platform give high AI density at attractive value.
Fast quotation, export experience
Tell us the workload and quantity and we normally return a configured quote within one working day, with export documentation and logistics arranged.
Configure your xFusion FusionServer G5500 V6
Send us your CPU, GPU, memory and storage requirements — or simply describe the workload — and we will come back with a tested build and a channel price.
xFusion FusionServer G5500 V6 questions
What is the FusionServer G5500 V6?
It is the previous-generation 4U accelerated server in xFusion’s flagship line, with two 3rd-Generation Intel Xeon Scalable processors, 32 DDR4 DIMMs, PCIe 4.0 and up to 8 FHFL dual-width GPU accelerators.
How many GPUs and which accelerators does it support?
It supports up to 8 full-height full-length dual-width cards, including NVIDIA A100/A800/A40-class accelerators, as well as FHFL and HHHL single-width cards, giving eight-card density at a lower platform cost than the V7.
Is it air-cooled or liquid-cooled?
It is air-cooled, using 6 or 8 hot-swap counter-rotating custom fans in N+1 redundancy and up to four 2000-3000 W Platinum/Titanium PSUs, so it runs in a standard rack with no liquid-cooling requirement.
Should I buy the G5500 V6 or the V7?
Choose the V6 when you want proven eight-card density on a cost-effective Ice Lake platform and PCIe 4.0 is sufficient. Choose the V7 when you need PCIe 5.0, DDR5, up to 10 dual-width GPUs and newer host performance for the fastest accelerators.
What storage does it support?
It holds 24 2.5/3.5-inch drives with a maximum of 4 NVMe SSDs, plus dual M.2 SSDs for boot, with hardware RAID 0/1/10/5/50/6/60 and supercapacitor protection.
What is a typical starting configuration?
A common value training build uses two Xeon Platinum 8380 processors, 512 GB of DDR4, eight A100/A800-class GPUs, NVMe boot/data drives and three OCP NICs with four high-wattage PSUs. We tailor each order to the workload.
Does it support GPUDirect and clustering?
Yes. One-click topology switchover between cascaded and balanced layouts, GPUDirect Storage/RDMA/P2P and up to three OCP 3.0 NICs let V6 nodes form an efficient accelerator pool for training or inference.
How do you test each G5500 V6 before shipment?
Every build goes through component verification, assembly and cable checks, full POST and firmware diagnostics, an extended high-utilization compute and GPU burn-in, storage and RAID validation, and a final independent QC sign-off. A test record accompanies the order and the 3-year warranty starts only after the unit passes, so no system ships untested.
What warranty, lead time and condition apply?
We supply the G5500 V6 as a brand-new server with a 3-year warranty, configured and burn-in tested before shipment. Quotations are normally provided within one working day and lead time depends on the CPU and GPU selected; export packaging and documentation are arranged for international orders. Read more on our warranty page.
WeChat
Scan the QR Code with wechat