AIGES is the core component of Athena Serving Framework. It is a general packaging tool for AI algorithm models and engines specially designed for AI capability developers. By integrating AIGES, you can quickly deploy AI algorithm models and engines, and host them on the Athena Serving Framework to use supporting auxiliary systems such as network, distribution strategy, and data processing.

Athena Serving Framework is committed to accelerating the AI ​​algorithm model and engine cloud service, and with the help of cloud native architecture, it provides multiple guarantees for the stability of cloud services. You don’t need to pay attention to the development, governance and operation and maintenance related to the underlying infrastructure and service, you can efficiently and safely deploy, upgrade, scale, operate and monitor models and engines.

Overall structure

characteristic:

• Support model reasoning into RPC service (Serving framework will be converted into HTTP service)
• Support C code reasoning
• Support Python code reasoning
• Supports once (non-streaming) reasoning and streaming reasoning
• Support configuration center, service discovery
• Support load balancing configuration
• Support HTTP/GRPC service