Speeding up Collective Communications Through Inter-GPU Re-Routing
Ranganath, Kiran, Abdolrashidi, AmirAli, Song, Shuaiwen Leon, Wong, Daniel
Published in IEEE computer architecture letters (01.07.2019)
Published in IEEE computer architecture letters (01.07.2019)
Get full text
Journal Article
MAPA: Multi-Accelerator Pattern Allocation Policy for Multi-Tenant GPU Servers
Ranganath, Kiran, Suetterlein, Joshua D., Manzano, Joseph B., Song, Shuaiwen Leon, Wong, Daniel
Published in SC21: International Conference for High Performance Computing, Networking, Storage and Analysis (14.11.2021)
Published in SC21: International Conference for High Performance Computing, Networking, Storage and Analysis (14.11.2021)
Get full text
Conference Proceeding
Energy Efficient Task Graph Execution Using Compute Unit Masking in GPUs
Chow, Marcus, Ranganath, Kiran, Lerias, Robert, Shanela Carodan, Mika, Wong, Daniel
Published in 2021 IEEE/ACM Redefining Scalability for Diversely Heterogeneous Architectures Workshop (RSDHA) (01.11.2021)
Published in 2021 IEEE/ACM Redefining Scalability for Diversely Heterogeneous Architectures Workshop (RSDHA) (01.11.2021)
Get full text
Conference Proceeding
MAPA: Multi-Accelerator Pattern Allocation Policy for Multi-Tenant GPU Servers
Ranganath, Kiran, Suetterlein, Joshua D, Manzano, Joseph B, Shuaiwen Leon Song, Wong, Daniel
Published in arXiv.org (07.10.2021)
Published in arXiv.org (07.10.2021)
Get full text
Paper
Journal Article
Toward a Holistic Performance Evaluation of Large Language Models Across Diverse AI Accelerators
Emani, Murali, Foreman, Sam, Sastry, Varuni, Xie, Zhen, Raskar, Siddhisanket, Arnold, William, Thakur, Rajeev, Vishwanath, Venkatram, Papka, Michael E., Shanmugavelu, Sanjif, Gandhi, Darshan, Zhao, Hengyu, Ma, Dun, Ranganath, Kiran, Weisner, Rick, Chen, Jiunn-yeu, Yang, Yuting, Vassilieva, Natalia, Zhang, Bin C., Howland, Sylvia, Tsyplikhin, Alexander
Published in 2024 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) (27.05.2024)
Published in 2024 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) (27.05.2024)
Get full text
Conference Proceeding