Active Member
|
[China]
Address: C1106,Jinyu Jiahua Building,Shangdi 3rd Street, Haidian District, Beijing 100085, P.R.China
Contact name:Cindy
Beijing Plink AI Technology Co., Ltd |
|
|
275 TOPS AI Medical Industrial IPC Embedded Computer NVIDIA Jetson AGX Orin
The NVIDIA Jetson industrial embedded pc (ORIN64-8F4E1) is with NVIDIA Jetson AGX Orin core module and the AI-enabled NVIDIA RTX 6000/A6000 GPU. Each reference application utilizes the necessary I/O components and drivers to move data and feed it into the processing pipeline on the RTX 6000 /A6000 GPU, providing a functionally tested data path setup, and it is the high performance embedded computing with NVIDIA Jetson AGX Orin 64G Module, the AI performance of Jetson AGX Orin module is 275 TOPS,The HD Audio-Video Subsystem uses a collection of functional blocks to off-load audio and video processing activities fromthe CPU complex, resulting in fast, fully concurrent, highly efficient operation.
This subsystem is comprised of the following:
The initial setting can be reset and restored
Maximum scalability 512g TF card memory
4G and WIFI module can be extended
| Optional expansion | 32GB ~ 1TB SSD storage |
| Processor | AGX Orin module |
| GPU | NVIDIA RTX 6000/A6000 |
| Memory | 64G |
| Weight | 5500g |
| Power requirements | 220V |
| Size | 335mm*257mm*146mm |
| Working temperature | -20~+65℃ |
NVIDIA Jetson Orin 64GB Technical Specifications
| Power | 15 W | 30 W | 50 W, and up to 60 W |
| CPU | 12-core Arm® Cortex®-A78AE v8.2 64-bit CPU 3MB L2 + 6MB L3 |
| GPU | 2048-core NVIDIA Ampere architecture GPU with 64 Tensor Cores |
| Vision and DNN accelerators | Up to 105 INT8 TOPS (Sparse, Deep Learning Inference) |
| Memory | 64 GB LPDDR5 |
| Storage | 64 GB eMMC 5.1 |
| AI Performance | 275 TOPS |
| Display | 1×: shared HDMI™ 2.1, eDP1.4, VESA® DisplayPort™ (DP) HBR3 |
| Ethernet | 1x GbE: RGMII 1× 10GbE: MGBE by XFI |
| PCIe | PCI Express 4.0 x1, x2, x4 and x8 Up to 2x Endpoint supported |
| USB | 3x: USB 3.2 4x: USB 2.0 |
| Other | UART, SPI, CAN, I2C, I2S, GPIOs |
| Module dimensions | 100.0 mm × 87.0 mm × 16.0 mm |
| Temperature range (at Thermal Transfer Plate (TTP) surface) | -25 °C to 80 °C |
| Operating humidity | 5% to 85% RH |
| Storage temperature | -25 °C to 80 °C |
| Storage humidity | 30% to 70% RH |
Note: See the NVIDIA Jetson AGX Orin Design Guide for details on the UPHY configurations supported. MGBE, USB 3.2, and PCIe share UPHY Lane.
NVIDIA Ampere GPU
Advanced GPU including CUDA cores, Ray-Tracing (RT) cores and 3rd Generation Tensor cores | Enhanced compute capability
2 GPC | 8 TPC | Up to 170 INT8 Sparse TOPS or 85 FP16 TFLOPS | Up to 5.32 FP32 TFLOPS or 10.649 FP16 TFLOPS (CUDA cores)
Arm Cortex-A78AE CPU
12× Arm Cortex-A78AE cores | 3 CPU clusters (4 cores/cluster) | 259 SPECint_rate2006
Arm® v8.2 (64-bit) | Symmetric multi-processing (SMP) | NEON SIMD | High-performance coherent interconnect fabric
Module Mechanical Drawing Side View
RTX 6000 /A6000 Specifications
| RTX 6000 | RTX A6000 | |
| NVIDIA CUDA Cores | 4608 | 10752 |
| NVIDIA Tensor Cores | 576 | 336 |
| GPU Memory | 24GB GDDR6 | 48GB GDDR6 |
| Memory interface | 384-bit | |
| Memory Bandwidth | Up to 672 GB/s | Up to 768 GB/s |
| Max Power Consumption | 295 W | 300 W |
| NVIDIA NVLink bandwidth | 100GB/s (bidirectional) | 112.5GB/s (bidirectional) |
| Error-correcting code (ECC) | Yes | |
