CCS Container Computing Service
Container Compute Service
It offers the best fully managed AI computing environment on the web, including development containers, task containers, and a wider range of generative AI services. The complete container management environment is perfectly integrated with storage space, providing a convenient, simple, and easy-to-understand user experience. With just a few clicks, you can quickly obtain a fully optimized AI computing environment from NGC.
CCS Container Computing Service
Taiwan AI Cloud's container computing service offers two container environments: development containers and task containers, which can help you:
- Rapidly deploy your development environment: Quickly edit programs using the built-in Jupyter Notebook and choose from a variety of AI frameworks.
- The container service automatically mounts the HFS high-speed archive system.
- It can deploy and execute a large number of AI calculations simultaneously, improving productivity.
Advantages
Abundant GPU resources meet the demands of large-scale AI training, reducing training time by nearly 501 TP3T and improving inference performance by nearly 2 times.
Employing a Kubernetes architecture and importing NVIDIA's optimized AI software stack enables rapid deployment of the working environment, reducing the deployment time for 30% container environments.
Large memory containers are easy to evaluate and convert to different specifications, and can be deployed and run simultaneously on a large number of AI computing containers to improve productivity.
A secure and fast storage system that provides the convenience of backup and deployment within the same working environment.
Simple user interface, easy access to computing services, flexible stacking and application.
Function
- It provides an advanced NVIDIA® Tesla V100 32GB SMX GPU computing accelerator with NVLink interconnect for multi-GPU setups.
- It is equipped with at least one V100 GPU + 4 CPU physical cores + 90GB main memory, and can call up to 8 GPUs for parallel computing to accommodate large-scale AI computing.
- Provides image files for various AI computing frameworks optimized from NVIDIA GPU Cloud. Includes common AI frameworks such as TensorFlow, Caffe, PyTorch, MXNet, Theano, and RAPIDS.
- It fully integrates with Jupyter Notebook, providing the ability to write programs, debug online, and perform calculations.
- Task-oriented containers allow you to specify the start time of computations and plan the amount of resources used by batch tasks appropriately.
- It provides a variety of tools, including UI, API, and CLI, to facilitate container management.
- You can log in to the container using SSH or Jupyter Notebook.
- The replication feature allows you to easily copy container environments, creating multiple identical containers for computation.
- Fully integrated with the HFS fully managed file system, eliminating worries about storage space.
Free Consultation Service
Contact Taiwan AI Cloud experts to learn about and start using the solution that suits you.