Video storage server reServer
GPUEDGEAI edge

Video storage server - reServer - Seeed Studio - GPU / EDGE / AI edge
Video storage server - reServer - Seeed Studio - GPU / EDGE / AI edge
Video storage server - reServer - Seeed Studio - GPU / EDGE / AI edge - image - 2
Video storage server - reServer - Seeed Studio - GPU / EDGE / AI edge - image - 3
Video storage server - reServer - Seeed Studio - GPU / EDGE / AI edge - image - 4
Add to favorites
Compare this product

Characteristics

Type
video storage, NVR, GPU, EDGE, AI edge
Mounting
desktop
Processor
NVIDIA, ARM Cortex-A78
Applications
industrial, networked camera, for IoT applications
Network
1GbE, 5G
Operating system
Linux, Linux® Ubuntu™
Other characteristics
gigabit Ethernet, SATA, NVMe, fanless, RS485, RS422, AI Inference
RAM capacity

16 GB

Min.: 4 GB

Max.: 16 GB

Description

Overview
The reServer Industrial Jetson™ Series delivers industrial-grade edge AI and NVR capabilities using NVIDIA Jetson Orin modules. Designed for continuous, fanless operation in wide temperature ranges, the series supports real-time video processing, multi-stream analytics and industrial I/O for integration with automation and surveillance systems.

Key Features
  • Fanless compact edge AI server: Powered by NVIDIA Jetson Orin modules up to 100 TOPS (Orin NX); optimized for deployment in harsh industrial environments with operating range -20 ~ 60°C.
  • Multi-stream video processing: Five GbE RJ45 ports (4 with 802.3af PSE) to support multiple camera streams and real-time analytics for surveillance and VMS applications.
  • Expandable storage: Two 2.5" SATA drive bays plus an M.2 2280 NVMe socket for flexible capacity and extended video retention.
  • Industrial connectivity: COM (RS232/RS422/RS485), DI/DO, CAN, USB3.1 and optional TPM2.0 for secure automation integration.
  • Hybrid wireless options: Supports optional 5G/4G/LTE and LoRaWAN modules via Nano SIM slot for remote or distributed installations.
  • Pre-installed JetPack and software ecosystem: Ready-to-run JetPack system with support for AI model deployment, CVEDIA-RT analytics packages and no-code workflow tools for rapid implementation.

Model Selection
Model | Processor | AI Performance | Memory | GPU Cores | DL Accelerator
J4012 | Orin NX 16GB | 100 TOPS | 16GB 128-bit LPDDR5 | 1024-core + 32 Tensor | 2x NVDLA v2
J4011 | Orin NX 8GB | 70 TOPS | 8GB 128-bit LPDDR5 | 1024-core + 32 Tensor | 1x NVDLA v2
J3011 | Orin Nano 8GB | 40 TOPS | 8GB 128-bit LPDDR5 | 1024-core + 32 Tensor | -
J3010 | Orin Nano 4GB | 20 TOPS | 4GB 64-bit LPDDR5 | 512-core + 16 Tensor | -

Additional Product Details
Smart storage reduces bandwidth and cloud costs by uploading only relevant events. The platform supports ready-made AI models (face, vehicle, people detection, crowd estimation) and enables no-code video analytics workflows via drag-and-drop tools. Documentation and best-practice guides cover data collection, annotation, model training/optimization and remote management.

Technical specifications
  • Jetson Module: Orin NX 16GB / Orin NX 8GB / Orin Nano 8GB / Orin Nano 4GB
  • AI Performance: 100 / 70 / 40 / 20 TOPS (model dependent)
  • GPU: NVIDIA Ampere-based GPU (1024-core with 32 Tensor Cores for Orin NX; 512-core with 16 Tensor Cores for Orin Nano 4GB)
  • CPU: Arm Cortex-A78AE (8-core for Orin NX; 6-core for other Orin variants)
  • Memory: 16GB/8GB/4GB LPDDR5 (model dependent)
  • DL Accelerator: 2x NVDLA v2 / 1x NVDLA v2 / -
  • Operating temperature: -20 ~ 60°C (with 0.7 m/s airflow)
  • Dimensions (W x D x H): 194.33 mm x 187 mm x 95.5 mm
  • Weight: 2.8 kg
  • Power: DC 12V–36V via 2-pin terminal block
  • Ethernet: 1x RJ45 GbE + 4x RJ45 GbE PoE (PSE 802.3af)
  • Storage: 2x 2.5" SATA bays + M.2 NVMe 2280
  • Industrial I/O: 4x DI/DO, RS232/RS422/RS485, CAN, USB3.1
*Prices are pre-tax. They exclude delivery charges and customs duties and do not include additional charges for installation or activation options. Prices are indicative only and may vary by country, with changes to the cost of raw materials and exchange rates.