Decentralized Generative AI Model Deployment Using Microservices
Sushant Jhingran, Nidhi Bansal, Rekha Chaturvedi, Ajeet Singh, Yojna Arora
Abstract
The microservices are continuously growing in domains of communication using cloud and AI. Access to microservice can be done using any respective cloud environment. Access microservices using cloud require multiple configurations. These configurations can be set up in Kubernetes or containerized environments to lightweight any application which increases the performance of microservices. Multiple requests can create a concurrent call, which causes slow response. Generative AI models can be used to solve concurrent calls. Generative AI used a modular approach for microservice communications with API gateway. This allows microservices to function in a variety of contexts, including cloud, edge devices, and containers. This process refers to a decentralized approach which can only be implemented in the case of microservices and provide improvement in latency along with bandwidth.