威九国际

Product Overview
High-performance AI Compute Pool

AI Compute Pool provides data center-level AI computing power resource pools, enabling user applications to transparently share and use AI accelerators on any server in the data center without modification.
It features elastic scalability and flexible billing, making your AI model training efficient and affordable.

Product Superiority
Rely on SenseTime's many years of experience in R&D and operation of super-large AI supercomputing clusters, building a training platform that understands AI training best.
  • 01Elastic scalability
  • 02High performance
  • 03Developer-friendliness
Elastic scalability
01Elastic scalability

Elastic scalability: Fully distributed deployment, connecting each node through an RDMA (IB/RoCE) or TCP/IP network to realize the elastic scalability of resource pools

High performance
02High performance

High performance: Ultimate hardware configuration and optimal scheduling optimization, bringing the most effective computing power supply for AI model training.

Developer-friendliness
03Developer-friendliness

Developer-friendliness: One-click solution to the problems of GPU/CPU ratio and multi-node multi-GPU model splitting in training models faced by AI developers, saving algorithm engineers a lot of valuable time

Elastic scalability
01
Elastic scalability

Elastic scalability: Fully distributed deployment, connecting each node through an RDMA (IB/RoCE) or TCP/IP network to realize the elastic scalability of resource pools

High performance
02
High performance

High performance: Ultimate hardware configuration and optimal scheduling optimization, bringing the most effective computing power supply for AI model training.

Developer-friendliness
03
Developer-friendliness

Developer-friendliness: One-click solution to the problems of GPU/CPU ratio and multi-node multi-GPU model splitting in training models faced by AI developers, saving algorithm engineers a lot of valuable time

01
/
03
Product Features
Model training、Task debugging、Task monitoring、Easy-to-use visualization tools.
  • Support lifecycle management of single-node and distributed training tasks in various frameworks
    Support lifecycle management of single-node and distributed training tasks in various frameworks

    Support TensorFlow and PyTorch training frameworks, MPI single-node and distributed training tasks, the configuration of various roles, the on-demand creation of training tasks, and the management of actions on tasks in various states; and provide multiple scheduling strategies to meet the requirements of various tasks for high efficiency and low cost;

  • Support RDMA networks
    Support RDMA networks

    Connect each node through an RDMA (IB/RoCE) or TCP/IP network to realize the elastic scalability of resource pools;

  • Support task observability
    Support task observability

    Provide monitoring and log services of various task resources and business metrics to meet the operation and maintenance requirements of algorithm engineers during the debugging process.

Application Scenarios
Based on deep learning methods; Computer vision application model training; Programmatic trading NLP model training; AI drug research.
  • 01AI training
  • 02Scientific computing
AI training
AI training
AI training makes use of high-performance GPU, high-speed network, and parallel file storage provided by high-performance AI Compute Pool, supports large-scale and high-concurrency task creation and scaling, and meets the demands of algorithm engineering for various computing power

Custom storages, networks, computing, and task schedulers for deep learning, combined with rich auxiliary debugging and visualization tools, bring efficient and developer-friendly deep learning training experience.

Scientific computing
Scientific computing
Gene sequencing performs sequencing and other processing on a large number of biological genomes to obtain genome information and data analysis results, helping solve problems in the fields of biology and medicine. The use of high-performance AI Compute Pool for new drug research and development can help researchers achieve fast concurrent processing of a large number of small molecule libraries.

Custom storages, networks, computing, and task schedulers for deep learning, combined with rich auxiliary debugging and visualization tools, bring efficient and developer-friendly deep learning training experience.

01AI training
02Scientific computing
AI training
AI training
AI training makes use of high-performance GPU, high-speed network, and parallel file storage provided by high-performance AI Compute Pool, supports large-scale and high-concurrency task creation and scaling, and meets the demands of algorithm engineering for various computing power

Custom storages, networks, computing, and task schedulers for deep learning, combined with rich auxiliary debugging and visualization tools, bring efficient and developer-friendly deep learning training experience.

Scientific computing
Scientific computing
Gene sequencing performs sequencing and other processing on a large number of biological genomes to obtain genome information and data analysis results, helping solve problems in the fields of biology and medicine. The use of high-performance AI Compute Pool for new drug research and development can help researchers achieve fast concurrent processing of a large number of small molecule libraries.

Custom storages, networks, computing, and task schedulers for deep learning, combined with rich auxiliary debugging and visualization tools, bring efficient and developer-friendly deep learning training experience.

Continuously update the whole line of products and insist on sincere communication and win-win cooperation

Help you achieve new breakthroughs in business with professional AI solutions and advanced AI products