Запити
Atlas 800
Tailored toward AI developers and data researchers, Huawei Atlas 800 provides end-to-end capabilities for data labeling, model generation, model training, and model inference services. With integrated software and hardware, this appliance reduces the technical entry requirements for AI applications and enables quick development and rollout of AI services for customers.
Huawei Atlas 800 AI appliance provides a Deep Learning Service (DLS) component as a one-stop deep learning platform. It integrates a large number of optimized network model algorithms to help users easily use deep learning technologies in a convenient and efficient manner, and provides users with model training, evaluation, and inference services through flexible scheduling and on-demand services.
Ready to work in 2 hours with a pre-installed AI development environment, underlying software library, and development framework. Automatic model generation with AutoDL, automatic hyperparameter tuning, and one-click model deployment.
Optimized AI environment based on a standard framework and programming environment. High-performance scheduling algorithms, improving resource utilization by over 15%.
Comprehensive management of resource utilization, health monitoring, and job scheduling. Easy-to-use WebUI, more intuitive and efficient than the Command Line Interface (CLI).
| Parameters | Model | |
| Model | Atlas 800 | |
| Cluster | Maximum nodes | 128 |
| Users | 1,000 | |
| Recommended Jobs (per user) | Training | 10 |
| Inference | 5 | |
| Visualization | 5 | |
| Code development | 4 | |
| Web UI and Services Processing Capability |
Maximum online users | 30 |
| Maximum concurrent requests | 30 per second | |
| Request processing latency | 3 seconds | |
| API Processing Capability | Maximum online users | 30 |
| Maximum concurrent requests | 30 per second | |
| Request processing latency | 3 seconds | |
| Hardware Specifications | AI accelerator card | Each full-width AI accelerator card: 8 full-height full-length AI accelerator cards (PCIe or NVLink) |
| Server | Each full-width 2-socket compute node: 2 Xeon® Scalable® processors, 24 DDR4 DIMMs |
|
| Hard drives | 2 x 2.5-inch SAS/SATA + 6 x 2.5-inch NVMe PCIe model: scalable to 8 x 3.5-inch SAS/SATA NVLink model: scalable to 8 x 2.5-inch SAS/SATA |
|
| RAID | RAID 0, 1, 10, 5, 50, 6, or 60 | |
| I/O | 4 x PCIe x16 LP + 2 x 10 GE LOM | |
| Power supply units | 4 hot-swappable 2,200W AC or 240V HVDC PSUs, with support for N+N redundancy | |
| Fan modules | 6 hot-swappable fan modules, with support for N+1 redundancy | |
| Temperature | 5°C to 35°C (41°F to 95°F) | |
| Dimensions (H x W x D) | 175 mm x 447 mm x 790 mm (6.89 in. x 17.60 in. x 31.10 in.) | |
Якщо ви вже є партнером компанії, натисніть тут, щоб отримати більше маркетингових ресурсів.
Натисніть тут, щоби перейти на сторінку партнерських організацій для перевірки стану запиту, управління замовленнями, отримання підтримки або додаткової інформації про партнерів компанії Huawei.
