Sotabase

Career

· UC Berkeley <i class="fa fa-arrow-right"></i> Duke

Publications (54)

Ansor : Generating High-Performance Tensor Programs for Deep Learning
USENIX Symposium on Operating Systems Design and Implementation · 2020
503
cited
This paper is included in the Proceedings of the 16th USENIX Symposium on Operating Systems Design and Implementation.
214
cited
TeraPipe: Token-Level Pipeline Parallelism for Training Large-Scale Language Models
International Conference on Machine Learning · 2021
155
cited
Understanding and Mitigating Packet Corruption in Data Center Networks
Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication · 2017
93
cited
Slim: OS Kernel Support for a Low-Overhead Container Overlay Network
Symposium on Networked Systems Design and Implementation · 2019
80
cited
Collie: Finding Performance Anomalies in RDMA Subsystems
Symposium on Networked Systems Design and Implementation · 2023
67
cited
Punica: Multi-Tenant LoRA Serving
Conference on Machine Learning and Systems · 2023
65
cited
Gallium: Automated Software Middlebox Offloading to Programmable Switches
Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication · 2020
63
cited
Understanding RDMA Microarchitecture Resources for Performance Isolation
Symposium on Networked Systems Design and Implementation · 2023
57
cited
An Online and Unified Algorithm for Projection Matrix Vector Multiplication with Application to Empirical Risk Minimization
International Conference on Artificial Intelligence and Statistics · 2023
41
cited
Dissecting Overheads of Service Mesh Sidecars
ACM Symposium on Cloud Computing · 2023
34
cited
Bypass Exponential Time Preprocessing: Fast Neural Network Training via Weight-Data Correlation Preprocessing
Neural Information Processing Systems · 2022
29
cited
Serving and Optimizing Machine Learning Workflows on Heterogeneous Infrastructures
Proceedings of the VLDB Endowment · 2022
29
cited
Training Overparametrized Neural Networks in Sublinear Time
arXiv.org · 2022
29
cited
Hoplite: efficient and fault-tolerant collective communication for task-based distributed systems
Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication · 2020
28
cited
High Velocity Kernel File Systems with Bento
USENIX Conference on File and Storage Technologies · 2020
25
cited
Subways: a case for redundant, inexpensive data center edge links
Conference on Emerging Network Experiment and Technology · 2015
25
cited
Automated Verification of Customizable Middlebox Properties with Gravel
Symposium on Networked Systems Design and Implementation · 2020
24
cited
RAIL: A Case for Redundant Arrays of Inexpensive Links in Data Center Networks
Symposium on Networked Systems Design and Implementation · 2017
22
cited
Remote Procedure Call as a Managed System Service
Symposium on Networked Systems Design and Implementation · 2023
21
cited
Sotabase
Danyang Zhuo | Researcher Profile | Sotabase | Sotabase