Multi-DNN Accelerators for Next-Generation AI Systems
This work tackles the problem of scalable and efficient AI processing for cloud and mobile systems, but it appears incremental as it builds on existing DNN accelerator concepts.
The paper addresses the challenge of increasing computational demands in AI systems by focusing on multi-DNN accelerator design to handle workloads from cloud services and mobile devices, aiming to maintain high performance as the number of DNNs grows.
As the use of AI-powered applications widens across multiple domains, so do increase the computational demands. Primary driver of AI technology are the deep neural networks (DNNs). When focusing either on cloud-based systems that serve multiple AI queries from different users each with their own DNN model, or on mobile robots and smartphones employing pipelines of various models or parallel DNNs for the concurrent processing of multi-modal data, the next generation of AI systems will have multi-DNN workloads at their core. Large-scale deployment of AI services and integration across mobile and embedded systems require additional breakthroughs in the computer architecture front, with processors that can maintain high performance as the number of DNNs increases while meeting the quality-of-service requirements, giving rise to the topic of multi-DNN accelerator design.