Rafique and Maurya receive Best Paper Award at High Performance Distributed Computing Conference

M. Mustafa Rafique, associate professor of computer science, and Avinash Maurya, a computer science Ph.D. student, received the Best Paper Award from the Association for Computing Machinery international symposium on High Performance Distributed Computing Conference. Their paper, titled "DataStates-LLM: Lazy Asynchronous Checkpointing for Large Language Models," addresses how to efficiently manage the saving and restoring of large AI models during training, ensuring minimal disruption and faster overall processing times.