Xin Ai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Xin Ai has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Sen Liu
Fudan University, Shanghai, China
4 shared publications - Yang Xu
Fudan University, Shanghai, China
4 shared publications - Zixuan Chen
China Telecom, Shanghai, China
4 shared publications - Yuanyi Zhu
Fudan University, Shanghai, China
3 shared publications - Zijian Li
Fudan University, Shanghai, China
3 shared publications - Lei Shi
Fudan University, China
1 shared publication
Latest Publications
- Open Access
- 16 June 2025
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 53 (1) , 145-147
NetJIT: Bridging the Gap from Traffic Prediction to Preknowledge for Distributed Machine Learning
Today's distributed machine learning (DML) introduces heavy traffic load, making the network one of the primary bottlenecks. To mitigate this bottleneck, existing state-of-the-art network optimization methods, such as traffic or topology engineering, are proposed to adapt to real-time traffic. However, current traffic measurement and prediction methods struggle to collect sufficiently fine-grained and accurate traffic patterns. This limitation impedes the ability of cutting-edge network optimization techniques to react agilely to the ever-changing traffic demands of DML jobs. This paper proposes NetJIT, a novel program-behavior-aware toolkit for foreseeing the traffic pattern of DML. To the best of our knowledge, this is the first work proposing the use of just-in-time (JIT) program analysis for real-time traffic measurement. In DML applications, communication behavior is primarily determined by the previously computed results. NetJIT leverages this characteristic to anticipate communication details by tracing and analyzing the data relations in the computation process. This capability enables the deployment of optimization strategies in advance. We deploy NetJIT in real-world network optimization for traffic preknowledge. Evaluation with the self-built testbed demonstrates that NetJIT can achieve up to about 97% less error of detecting communication events compared with other methods. Simulations with real-world DML workloads further illustrate that NetJIT enables more precise network optimization, leading to approximately 50% better network performance w.r.t the metrics including average iteration time, throughput, and average packet delay.
- 16 June 2025
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 53 (1) , 145-147
Open AccessNetJIT: Bridging the Gap from Traffic Prediction to Preknowledge for Distributed Machine Learning
Today's distributed machine learning (DML) introduces heavy traffic load, making the network one of the primary bottlenecks. To mitigate this bottleneck, existing state-of-the-art network optimization methods, such as traffic or topology engineering, are proposed to adapt to real-time traffic. However, current traffic measurement and prediction methods struggle to collect sufficiently fine-grained and accurate traffic patterns. This limitation impedes the ability of cutting-edge network optimization techniques to react agilely to the ever-changing traffic demands of DML jobs. This paper proposes NetJIT, a novel program-behavior-aware toolkit for foreseeing the traffic pattern of DML. To the best of our knowledge, this is the first work proposing the use of just-in-time (JIT) program analysis for real-time traffic measurement. In DML applications, communication behavior is primarily determined by the previously computed results. NetJIT leverages this characteristic to anticipate communication details by tracing and analyzing the data relations in the computation process. This capability enables the deployment of optimization strategies in advance. We deploy NetJIT in real-world network optimization for traffic preknowledge. Evaluation with the self-built testbed demonstrates that NetJIT can achieve up to about 97% less error of detecting communication events compared with other methods. Simulations with real-world DML workloads further illustrate that NetJIT enables more precise network optimization, leading to approximately 50% better network performance w.r.t the metrics including average iteration time, throughput, and average packet delay.
- Open Access
- 9 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 145-147
NetJIT: Bridging the Gap from Traffic Prediction to Preknowledge for Distributed Machine Learning
- 9 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 145-147
Open AccessNetJIT: Bridging the Gap from Traffic Prediction to Preknowledge for Distributed Machine Learning
- Open Access
- 6 March 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Measurement and Analysis of Computing Systems
- Vol. 9 (1) , 1-25
NetJIT: Bridging the Gap from Traffic Prediction to Preknowledge for Distributed Machine Learning
Today's distributed machine learning (DML) introduces heavy traffic load, making the interconnection network one of the primary bottlenecks. To mitigate this bottleneck, existing state-of-the-art network optimization methods, such as traffic or topology engineering, are proposed to adapt to real-time traffic. However, current traffic measurement and prediction methods struggle to collect sufficiently fine-grained and accurate traffic patterns. This limitation impedes the ability of cutting-edge network optimization techniques to react agilely to the ever-changing traffic demands of DML jobs. This paper proposes NetJIT, a novel program-behavior-aware toolkit for accurately foreseeing the traffic pattern of DML. To the best of our knowledge, this is the first work proposing the use of just-in-time (JIT) program analysis for real-time traffic measurement. In DML applications, communication behavior is primarily determined by the previously computed results. NetJIT leverages this characteristic to anticipate communication details by tracing and analyzing the data relations in the computation process. This capability enables the deployment of optimization strategies in advance. We deploy NetJIT in real-world network optimization for traffic preknowledge. Evaluation with the self-built testbed prototype demonstrates that NetJIT can achieve up to about 97% less error of detecting communication events compared with other methods. Simulations with real-world DML workloads further illustrate that NetJIT enables more precise network optimization, leading to approximately 50% better network performance w.r.t the metrics including average iteration time, throughput, and average packet delay.
- 6 March 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Measurement and Analysis of Computing Systems
- Vol. 9 (1) , 1-25
Open AccessNetJIT: Bridging the Gap from Traffic Prediction to Preknowledge for Distributed Machine Learning
Today's distributed machine learning (DML) introduces heavy traffic load, making the interconnection network one of the primary bottlenecks. To mitigate this bottleneck, existing state-of-the-art network optimization methods, such as traffic or topology engineering, are proposed to adapt to real-time traffic. However, current traffic measurement and prediction methods struggle to collect sufficiently fine-grained and accurate traffic patterns. This limitation impedes the ability of cutting-edge network optimization techniques to react agilely to the ever-changing traffic demands of DML jobs. This paper proposes NetJIT, a novel program-behavior-aware toolkit for accurately foreseeing the traffic pattern of DML. To the best of our knowledge, this is the first work proposing the use of just-in-time (JIT) program analysis for real-time traffic measurement. In DML applications, communication behavior is primarily determined by the previously computed results. NetJIT leverages this characteristic to anticipate communication details by tracing and analyzing the data relations in the computation process. This capability enables the deployment of optimization strategies in advance. We deploy NetJIT in real-world network optimization for traffic preknowledge. Evaluation with the self-built testbed prototype demonstrates that NetJIT can achieve up to about 97% less error of detecting communication events compared with other methods. Simulations with real-world DML workloads further illustrate that NetJIT enables more precise network optimization, leading to approximately 50% better network performance w.r.t the metrics including average iteration time, throughput, and average packet delay.
- Open Access
- 28 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 65-66
Augmenting Distributed AI Training with Loss-tolerant Transmission
- 28 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 65-66
Open AccessAugmenting Distributed AI Training with Loss-tolerant Transmission