News

Six Nodes, Six GPUs, One Chassis? TFNEW ZN-6A

TFNEW launched the ZN-6A, a 6-node multi-node server where each Core i9/i7 node has its own fan-cooled GPU, PCIe 5.0/4.0 x16 slots and independent 1300W CRPS power, for cloud gaming, edge and channel-analytics platforms.

Six Nodes, Six GPUs, One Chassis? TFNEW ZN-6A

Six Nodes, Six GPUs, One Chassis? TFNEW ZN-6A

TL;DR: TFNEW launched the ZN-6A, a 6-node multi-node server, on September 21, 2026. One chassis holds six independent nodes — each a 14th/13th Gen Intel Core i9/i7 with DDR5 memory, its own set of PCIe 5.0/4.0 x16 slots and a fan-cooled GPU card — with independent power per node, for cloud gaming, cloud computing and datacenter deployments that need GPU-accelerated node density. Get specs and pricing: sales@tfnew.com.

Core Specifications (with Plain-Language Terms)

The ZN-6A is a multi-node server: six independent servers in one chassis (264 mm tall, 800 mm deep), each node running a 14th/13th Gen Intel Core i9/i7 processor on an Intel B760 chipset. Memory is 24 DDR5 slots across the chassis (up to 7800 MT/s). The ZN-6A's differentiator versus plain multi-node chassis: GPU support per node — six fan-cooled graphics cards in total, with 6x PCIe 5.0 x16 and 6x PCIe 4.0 x16 slots chassis-wide, so each node can carry its own accelerator. Storage is 4x SATA 3.0 plus 2x M.2 per node. Power is 6x CRPS 1300 W supplies — one per node — so each node's power domain is its own (independent power supply and operation). Cooling is 2x6 high-performance fans; networking is integrated Realtek 2.5GbE (2.5 Gbps / 1 Gbps / 100 Mbps) per node. A multi-level cascade architecture supports statistics and analysis of user and channel information across the cluster.

Specs at a Glance

Item ZN-6A Value (per node / chassis)
Form factor Multi-node chassis, 264 H × 442 W × 800 D mm, 6 nodes
Processor 6x 14th/13th Gen Intel Core i9/i7, Intel B760
Memory 24x DDR5 total (4 per node), up to 7800 MT/s
GPU 6x fan-cooled graphics cards (PCIe 5.0 / 4.0 x16)
PCIe 6x PCIe 5.0 x16 + 6x PCIe 4.0 x16 chassis-wide
Storage 4x SATA 3.0 + 2x M.2 per node
Power 6x CRPS 1300 W, independent per node
Cooling 2x6 high-performance fans
Network Realtek 2.5GbE per node (2.5G / 1G / 100M)

Why GPU Per Node Is the Multi-Node Twist

Plain multi-node servers give you nodes; they leave GPU density to the rack. The ZN-6A gives each of its six nodes its own accelerator path — a fan-cooled graphics card behind per-node PCIe 5.0/4.0 x16 slots — which is exactly the profile cloud-gaming and edge-compute operators need: individual GPU workloads, isolated from each other, in one chassis. Independent 1300 W CRPS per node keeps the isolation physical: a power event on node three cannot take node four down. That combination — GPU per node, power per node, management per node — is what separates a GPU multi-node box from a row of separate servers.

Where It Fits

  • Cloud gaming: per-node GPU for game-session density in one chassis
  • Cloud computing: accelerated single-socket nodes with private power domains
  • Edge and regional POPs: GPU-compute nodes in shallow 800 mm racks
  • Channel analytics platforms: cascade architecture for user/channel statistics

Why the Numbers Matter

  • 6 nodes, each GPU-capable: six isolated accelerated workloads per chassis
  • 6x CRPS 1300 W, one per node: independent power failure domains
  • PCIe 5.0 + 4.0 x16 per node: accelerator and controller headroom everywhere
  • 24x DDR5 at 7800 MT/s: fast memory for game and VM instances
  • Cluster cascade management: node-level statistics and monitoring as standard

What TF NEW Says

"Multi-node multiplied by GPU is a different product than either alone," said a TF NEW product manager. "The ZN-6A gives six Core i9/i7 nodes, each with its own graphics card and its own 1300 W power feed, in one chassis. Cloud-gaming and edge operators get six isolated accelerated machines in the space of one — with power isolation built in."

Quality and Availability

Every ZN-6A runs 48 hours of burn-in testing across all six nodes and their GPU paths before shipping, with per-node power domains validated in the factory. It ships deployment-ready with cascade management documentation and cluster setup guides.

Pricing and Availability

The ZN-6A is available now. Pricing varies with processor, memory, GPU and storage configuration. Contact sales@tfnew.com or use the request-a-quote form at www.tfnew.com/contact for volume pricing and lead time.

About TF NEW

TF NEW (TF New Technology) is an enterprise AI infrastructure provider based at www.tfnew.com. It designs and delivers private AI, compute, networking, storage, cooling and edge systems as deployment-ready solutions — AI servers, AI workstations, edge AI, liquid cooling and enterprise rack servers — for AI factories, datacenters and edge projects worldwide.

Media Contact

TF NEW | sales@tfnew.com | www.tfnew.com/contact

FAQ

  1. What is the ZN-6A? A multi-node server with six independent nodes, each able to host its own fan-cooled graphics card.
  2. How many GPUs can it hold? Six — one per node, on per-node PCIe 5.0/4.0 x16 slots.
  3. What processors does each node run? 14th/13th Gen Intel Core i9/i7 on an Intel B760 chipset.
  4. How is power handled? Six independent CRPS 1300 W supplies, one per node.
  5. What workloads is it for? Cloud gaming, cloud computing, edge/POP deployments and channel-analytics platforms.