How TenClass saved 80% on GPU costs with TensorFusion?
TenClass using TensorFusion to save 80% on GPU costs
Background
TenClass is a vocational education company based in Shenzhen, China, providing digital literacy and professional skills courses to help users enhance their digital skills.
Problem
To provide a convenient and high-quality hands-on learning environment, TenClass developed a virtual machine manager mvisor, but in the construction of hands-on learning environments for AI drawing and AI video courses, due to the need for direct GPU access in online laboratories, due to the immaturity of GPU virtualization technology, providing a dedicated AI drawing environment for each learner is very difficult and expensive.
Solution
After learning that TensorFusion can achieve GPU virtualization and pooling, after a period of evaluation and implementation, TenClass became the first customer of TensorFusion.
