TopoOpt: Co-optimizing Network Topology and Parallelization Strategy for Distributed Training Jobs W Wang, M Khazraee, Z Zhong, M Ghobadi, Z Jia, D Mudigere, Y Zhang, ... 20th USENIX Symposium on Networked Systems Design and Implementation (NSDI …, 2023 | 66* | 2023 |
Adapting {TCP} for reconfigurable datacenter networks MK Mukerjee, C Canel, W Wang, D Kim, S Seshan, AC Snoeren 17th USENIX Symposium on Networked Systems Design and Implementation (NSDI …, 2020 | 26 | 2020 |
IOI: In-network optical inference Z Zhong, W Wang, M Ghobadi, A Sludds, R Hamerly, L Bernstein, ... Proceedings of the ACM SIGCOMM 2021 Workshop on Optical Systems, 18-22, 2021 | 14 | 2021 |
How to Build Low-cost Networks for Large Language Models (without Sacrificing Performance)? W Wang, M Ghobadi, K Shakeri, Y Zhang, N Hasani arXiv preprint arXiv:2307.12169, 2023 | 11* | 2023 |
Time-division TCP for reconfigurable data center networks SS Chen, W Wang, C Canel, S Seshan, AC Snoeren, P Steenkiste Proceedings of the ACM SIGCOMM 2022 Conference, 19-35, 2022 | 11 | 2022 |
Efficient Direct-Connect Topologies for Collective Communications L Zhao, S Pal, T Chugh, W Wang, J Fantl, P Basu, J Khoury, ... arXiv preprint arXiv:2202.03356, 2022 | 7* | 2022 |
In-network optical inference M Ghobadi, Z Zhong, W Weiyang, LSB Bernstein, A Sludds, R Hamerly, ... US Patent App. 18/561,985, 2024 | | 2024 |
Rail-only: A Low-Cost High-Performance Network for Training LLMs with Trillion Parameters W Wang, M Ghobadi, K Shakeri, Y Zhang, N Hasani | | |