We use cookies to improve our site and your experience. By using this website, you accept our privacy terms.

CDN

Benefits

Diverse Selection

A variety of mid to high end GPU cards such as K80, M60, P100, V100 are used to meet the computing needs of different application scenarios.

Cost-effective

Adopting mainstream GPU hardware and technology, supporting on-demand purchase, a very cost-effective solution.

Easy To Use

Provides detailed step-by-step installation guide of GPU drivers, along with authorization server for vGPU, saving users’ time on deploying GPU environment.

Easy Management

Batch creation, flexible network configuration, and custom private images are supported, as well as supporting cloud monitoring to manage the operation of GPU cloud servers.

Features

  • Ultra Powerful Computing

    Ultra high GPU computing power, easy to cope with high concurrency and massive computing scenarios.

    • High-spec computing nodes: each GPU cloud server has its own physical GPU card to ensure stable and reliable performance.
    • Diverse selection of GPU cards: a variety of high-performance computing cards for users to choose from. Each GPU cloud server can be equipped with 1, 2, 4, or 8 GPU cards.

  • Virtual Private Network (VPC)

    Build a secure and isolated network environment with support for custom network types, subnetting and other operations

  • Flexible Cloud Hard Disk

    Supporting SSD cloud hard disk or performance cloud hard disk for expansion of storage resources

  • Image

    Rapid creation of CPU cloud server with easy deployment

    • Integrated GPU driver: integrating the latest GPU drivers and CUDA to address underlying compatibility issues.
    • System image: public images of multiple releases are supported, users may choose according to their needs.
    • Custom image: users can produce custom images based on the deployed GPU cloud server system drive to quickly create more GPU cloud servers.

  • Monitoring Service

    Provide fine-grained GPU card status monitoring and alert mechanism, including GPU utilization, memory utilization, and temperature.

Scenarios

  • Deep Learning
  • Scientific Computing
  • Image/Video Processing

  • Single GPU cloud server provides in maximum 125TFLOPS single-precision floating-point computing capability, which can easily meet the computing power requirements in deep learning scenarios, and supports deep learning training and inference, image recognition, image content identification, speech recognition and other services.

  • The integration of Wangsu cloud load balancing, object storage, cloud security and CDN helps to quickly build a powerful deep learning and training system and shorten the training cycle.

  • GPU cloud servers provide high-performance parallel computing capabilities, where a single GPU cloud server is capable of in maximum 62TFLOPS double-precision floating-point computing.

  • Ideal for collision simulation, computational finance, genetic engineering, weather prediction and other fields that require high precision computing capabilities of the server.

  • Provide professional computing power for video and graphics processing.

  • Ideal for film and television rendering production, 4K/8K HD live streams, special effects processing, 3D industrial design, 3D demonstration, engineering, etc.