Skip to content

Enterprise servers. Designed in the USA.

The Boreon zNAS, zServe, and Phoenix product lines: all-flash storage, compute servers, and AMD EPYC AI infrastructure. Built for ministries, businesses, and AI datacenters.

Boreon zNAS: All-Flash Storage Servers

10GbE all-flash NAS for SMB/CIFS, iSCSI, AFP Time Machine, and NFS. ZFS ARC/L2ARC in-memory caching, Active Directory sync, FreeBSD OS. Designed and assembled in the USA.

Boreon zNAS 10 all-flash storage server

zNAS 10

10TB all-flash · 1U rackmount · 10GbE

Small Business · 1–10 Users

  • 10TB usable: 6× 1.9TB Intel MLC Enterprise SSDs
  • Intel Xeon E5 3.5 GHz Quad-Core
  • 32GB (4× 8GB) Samsung ECC registered memory
  • 1U rackmount · 8× 2.5" SSD hot-swap bays · dual 700W PSU
  • 1 GbE: Intel dual-port RJ45 · 10 GbE: Chelsio dual-port SFP+
  • SMB/CIFS · AFP Time Machine · iSCSI · NFS · ZFS ARC
Boreon zNAS 50 enterprise all-flash storage

zNAS 50

50TB all-flash · 2U rackmount · 10GbE

Enterprise · 10–50 Users

  • 50TB usable: 20× 2.5TB Intel MLC Enterprise SSDs
  • Intel Xeon E5 5 GHz Hex-Core Server · 64GB ECC memory
  • 2U rackmount · 24× 2.5" SSD hot-swap bays · dual 1200W PSU
  • 10 GbE: Chelsio dual-port 10 GbE SFP+ (optional)
  • SMB/CIFS · AFP · iSCSI · NFS · ZFS with ARC + L2ARC
  • Active Directory credential sync · snapshot scheduling

Boreon zServe: All-Flash Compute Servers

Enterprise compute from small business to AI-scale. Intel Xeon · all-flash or high-density hybrid · Windows Server, VMware, Ubuntu, CentOS.

Boreon zServe 10

zServe 10

All-flash compute · 1U

Small Business · 1–10 Users

  • 800GB all-flash array: 2× 400GB Intel DC MLC SSDs
  • Intel Xeon E5 2.5 GHz 6-Core Turbo
  • 16GB (2× 8GB) ECC registered memory
  • 1U · dual 200W PSU · 1 GbE + optional 10 GbE
Boreon zServe 50

zServe 50

64TB hybrid storage · 2U

Enterprise · 2–50 Users

  • 64TB hybrid: 8× WD Red NAS + hot-swap SSDs
  • Intel Xeon E5 3.0 GHz Turbo Boost Dual-Controller
  • 640GB ECC memory · dual Intel Xeon scalable
  • 2U · dual redundant PSU · 10 GbE SFP+
Boreon zServe 100

zServe 100

120TB hybrid storage · 2U

Enterprise · 2–500 Users

  • 120TB hybrid: high-density hot-swap SSD array
  • Dual Intel Xeon scalable · 384GB ECC
  • Intel Tolapai SSD Write Cache · dual controller
  • 2U · Fans: 6× 40mm · 10 GbE x4 RJ45 + SFP+

Boreon Phoenix: AMD EPYC AI & HPC Systems

The Phoenix line is Boreon's high-performance compute platform. It is built for AI accelerator fleets, machine learning workloads, and HPC environments that demand maximum core density and memory bandwidth.

AMD EPYC multi-socket architecture delivers up to 512 cores per rack node with NUMA-aware memory topology. Paired with our KWAM erasure-coding layer, Phoenix nodes form self-healing, fault-tolerant AI data clusters with six-nines durability design targets.

  • AMD EPYC multi-socket compute · up to 512 cores per node
  • High-bandwidth DDR4/DDR5 ECC registered memory
  • PCIe Gen 4/5 GPU and accelerator expansion
  • KWAM-integrated: erasure-coded, self-healing storage fabric
  • Custom rack configurations for AI datacenter deployments
Boreon Phoenix AMD EPYC rack Boreon Phoenix AMD EPYC 2U node

OS compatibility

All Boreon server products are validated for Windows Server, VMware vSphere, Ubuntu Server, and CentOS/RHEL. Custom OS configurations and bare-metal deployments available on request.

Boreon Hosting

Where Boreon runs its own cloud. Our Sovereign servers, our demo AI platforms, and every Boreon-hosted local open AI model run on this infrastructure.

Infomaniak Public Cloud: Swiss-hosted, enterprise-grade infrastructure
Infomaniak Public Cloud: Swiss-hosted, enterprise-grade infrastructure. · Boreon infrastructure runs on Infomaniak Public Cloud, with Swiss data sovereignty, 24/7 uptime, and enterprise-grade security. It is ISO-27001 certified and 100% renewable-energy, with zero data resale and full GDPR compliance.
  • Sovereign servers: dedicated instances Boreon operates for customers who want the platform run for them.
  • Demo AI platforms: the live trial environments behind ai.boreon.com.
  • Boreon local open AI models: open-weight models we host ourselves, so a prompt never leaves infrastructure we control.

Book a free discovery call.

Tell us about your data, your migration, or your network. We'd love to serve you.