Cernis & Donati series – MIL-rugged SFF edge AI compute systems
- Technology
- GPGPU-HPC
- Partner
- One Stop Systems
The Cernis and Donati rugged embedded systems from OSS deliver breathtaking AI inferencing capabilities inside MIL-grade, small-form-factor platforms built for the extreme edge. These systems empower defence, aerospace, industrial, and autonomous applications with datacentre power at the tactical edge.
With up to 6.6 PetaOPS of AI compute, 150GB/s data-fabric bandwidth, and advanced PCIe switching, they’re built to accelerate sensor-rich, mission-critical workloads — all while thriving in punishing conditions including shock, vibration, dust, and temperature extremes.
If your application demands ultra-low latency processing, secure on-prem AI compute, and rugged reliability, this platform is an absolute powerhouse.
Range features
A high level overview of what this range offers
-
Up to 6.6 PetaOPS AI performance Run next-generation AI models, high-speed sensor fusion, autonomy stacks, and advanced analytics right at the edge — no cloud dependency required.
-
150GB/s high-speed data fabric Blazing-fast data movement ensures real-time ingest and processing of multi-sensor streams, including radar, EO/IR, LiDAR, SIGINT, and mission systems.
-
Advanced PCIe switching architecture Supports scalable, multi-accelerator configurations and deterministic, low-latency performance across complex embedded workloads.
-
MIL-rugged construction Designed for battlefield and aerospace environments — resistant to shock, vibration, dust, humidity, extreme temps, and electromagnetic interference.
-
Compact SFF deployability Despite its immense power, the SFF design enables easy integration into UAVs, UGVs, aircraft pods, ground vehicles, and mobile command systems.
-
Edge-optimised for AI inference Perfect for object detection, target tracking, SLAM, navigation, digital SIGINT, real-time surveillance, and autonomous decision-making.
-
Scalable, modular architecture Flexible accelerator options allow mission-specific configurations — from power-optimised setups to maximum-throughput AI engines.
-
Mission-ready reliability Built for 24/7 operation in remote or hostile environments where downtime is not an option.
Downloads
for Cernis & Donati series - MIL-rugged SFF edge AI compute systems
What’s in this range?
All the variants in the range and a comparison of what they offer
| Specification | Cernis | Donati |
|---|---|---|
Dimensions | 14.9” H × 6.51” W × 3.15” D | 8.5” H × 4.8” W × 3.15” D |
AI Performance | 40 TOPS | 275 TOPS |
GPU | 1024-core NVIDIA Ampere GPU with 32 Tensor Cores | 2048-core NVIDIA Ampere GPU with 64 Tensor Cores |
GPU Max Frequency | 625 MHz | 1.3 GHz |
CPU | 6-core Arm® Cortex®-A78AE v8.2 64-bit CPU (1.5 MB L2 + 4 MB L3) | 12-core Arm® Cortex®-A78AE v8.2 64-bit CPU (3 MB L2 + 6 MB L3) |
CPU Max Frequency | 1.5 GHz | 2.2 GHz |
System Memory | 8 GB 128-bit LPDDR5 (68 GB/s) | 64 GB 256-bit LPDDR5 (204.8 GB/s) |
Boot Memory | 64 GB eMMC 5.1 | 64 GB eMMC 5.1 |
Network Controllers | 2× GbE | 2× GbE |
Chassis | Milled Aluminum, Flange Mount | Milled Aluminum, Flange Mount |
Weight | 19.1 lbs | 10.55 lbs |
Environment — Temperature | Operating: −40°C to 85°C @ 10,000 ft | Operating: −40°C to 85°C @ 10,000 ft |
Environment — Humidity | Operating: 0%–100% condensing RH | Operating: 0%–100% condensing RH |
Shock | ±10 g, 11 ms, half-sine pulse, 3 shocks per axis | ±10 g, 11 ms, half-sine pulse, 3 shocks per axis |
Vibration | 4.5 Grms, 10–2000 Hz | 4.5 Grms, 10–2000 Hz |
Agency Compliance | Designed to conform to: | Designed to conform to: |
