DVABatch: Diversity-aware Multi-Entry Multi-Exit Batching for Efficient Processing of DNN Services on GPUs

TitleDVABatch: Diversity-aware Multi-Entry Multi-Exit Batching for Efficient Processing of DNN Services on GPUs
Publication TypeConference Paper
Year of Publication2022
AuthorsCui W, Zhao H, Chen Q, Wei H, Li Z, Zeng D, Li C, Guo M
Conference Name2022 USENIX Annual Technical Conference (USENIX ATC 22)
Date Published07/2022
PublisherUSENIX Association
Conference LocationCarlsbad, CA
ISBN Number978-1-939133-29-10
URLhttps://www.usenix.org/conference/atc22/presentation/cui