Google Cloud team
Conducted research for On-Chip-Network of SoC in area of routing algorithms and Quality of Service as part of the DOE FastForward2 project
Designed and implemented an assembly methodology framework to drive simulations in the exascale computing research domain to serve the DOE FastForward project
Mingchao Yu*, Zhifeng Lin*(Equal Contributions), Krishna Narra, Songze Li, Youjie Li, Nam Sung Kim, Alexander Schwing, Murali Annavaram, Salman Avestimehr. GradiVeQ: Vector Quantization for Bandwidth-Efficient Gradient Aggregation in Distributed CNN training. In the 31st Annual Conference on Neural Information Processing Systems (NeurIPS), Montreal, Quebec, Canada, December 2018
View Paper View VideoKrishna Giri Narra*, Zhifeng Lin*(Equal Contributions), Mehrdad Kiamar, Salman Avestimehr, Murali Annavaram. Slack Squeeze Coded Computing for Adaptive StragglerMitigation. To appear in the International Conference for High Performance Computing Networking, Storage, and Analysis 2019 (SC19), Denver, CO, November 2019
Jieming Yin, Zhifeng Lin, Onur Kayiran, Matthew Poremba, Muhammad Shoaib Bin Altaf, Natalie Enright Jerger, Gabriel H. Loh. Modular Routing Design for Chiplet-based Systems. In the Proceedings of the 45th International Symposium on Computer Architecture (ISCA), Los Angeles, CA, June 2018
View Paper View VideoKrishna Giri Narra, Zhifeng Lin, Ganesh Ananthanarayanan, Salman Avestimehr, Murali Annavaram. Collage Inference: Achieving low tail latency during distributed image classification using coded redundancy models. CodML workshop (Workshop on Coding Theory For Large-scale Machine Learning) at the 36th International Conference on Machine Learning (ICML), Long Beach, CA, June 2019
View PaperZhifeng Lin, Yasuko Eckert, Gabriel H. Loh. Unified and Flexible Methodology for System Specification and Evaluation. Workshop on Modeling & Simulation of Systems and Application 2016 (ModSim), University of Washington, Seattle, WA, August 2016
View Presentation