Yijun Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yijun Li has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Jiawei Huang
Central South University
14 shared publications - Zhaoyi Li
Central South University, Changsha, China
11 shared publications - Jianxin Wang
School of Computer Science and Engineering, Central South University, Changsha, Hunan, China
10 shared publications - Jingling Liu
Central South University, Changsha, China
6 shared publications - Wanchun Jiang
Central South University
5 shared publications - Jin Ye
School of Computer and Electronic Information, Guangxi University, Nanning, China
3 shared publications
Latest Publications
- Open Access
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 1019-1029
PipeTree: Decision Tree Partitioning for Efficient Inference on Programmable Switches
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 1019-1029
Open AccessPipeTree: Decision Tree Partitioning for Efficient Inference on Programmable Switches
- 19 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 2939-2953
FAR: Fast and Accurate Rate Control for Lossless Datacenter Networks
In recent years, end-to-end congestion control algorithms or flow pausing mechanisms are proposed to achieve high throughput and low latency in datacenter networks. However, prior end-to-end congestion control works without complex signals fail to achieve fast convergence to a stable equilibrium state and effectively handle the transient congestion, while existing flow pausing mechanisms are decoupled from congestion control, which leads to long convergence time after transient states and incomplete queue elimination in equilibrium states. To address these issues, we present FAR, a rate control protocol that combines the advantages of flow pausing and congestion control. At its heart, FAR couples the bandwidth-estimation-based congestion control and the end-to-end flow pausing mechanisms. After flow pausing, FAR quickly explore the available bandwidth with a binary-search probe to achieve high throughput and low latency. Meanwhile, FAR employs a probe staggering mechanism to address the queue oscillation issue in high-concurrency scenarios. We implement the prototype of FAR using DPDK. Extensive evaluation results demonstrate that our protocol achieves accurate bandwidth estimation and reduces the tail flow completion time (FCT) by up to 67% compared with the state-of-the-art designs.
- 19 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 2939-2953
FAR: Fast and Accurate Rate Control for Lossless Datacenter Networks
In recent years, end-to-end congestion control algorithms or flow pausing mechanisms are proposed to achieve high throughput and low latency in datacenter networks. However, prior end-to-end congestion control works without complex signals fail to achieve fast convergence to a stable equilibrium state and effectively handle the transient congestion, while existing flow pausing mechanisms are decoupled from congestion control, which leads to long convergence time after transient states and incomplete queue elimination in equilibrium states. To address these issues, we present FAR, a rate control protocol that combines the advantages of flow pausing and congestion control. At its heart, FAR couples the bandwidth-estimation-based congestion control and the end-to-end flow pausing mechanisms. After flow pausing, FAR quickly explore the available bandwidth with a binary-search probe to achieve high throughput and low latency. Meanwhile, FAR employs a probe staggering mechanism to address the queue oscillation issue in high-concurrency scenarios. We implement the prototype of FAR using DPDK. Extensive evaluation results demonstrate that our protocol achieves accurate bandwidth estimation and reduces the tail flow completion time (FCT) by up to 67% compared with the state-of-the-art designs.
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 1838-1850
SIM: Accelerating Distributed DNN Training by Exploring Gradient Similarity
Synchronous stochastic gradient descent (SSGD) has been widely used in distributed deep learning. However, since the local gradients need to be shared among workers at every iteration, SSGD performance is significantly influenced by network bottlenecks caused by either heterogeneous environment or bandwidth contention. To solve this problem, asynchronous parallel (ASP) strategy allows each worker to update parameters independently without synchronization, while suffering from accuracy loss and convergence inefficiency. In this paper, we propose a novel similarity-based synchronization scheme called SIM, which mitigates the impact of network bottlenecks and ensures convergence efficiency. Specifically, SIM reduces the number of aggregation workers based on the gradient similarity between global and local gradients, therefore shrinking the waiting time for the stragglers. We provide a theoretical analysis of convergence efficiency and conduct large-scale testbed experiments on CIFAR-10 and SQUAD dataset. The experimental results show that SIM reduces the convergence time of four classical deep learning models by up to 40%.
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 1838-1850
SIM: Accelerating Distributed DNN Training by Exploring Gradient Similarity
Synchronous stochastic gradient descent (SSGD) has been widely used in distributed deep learning. However, since the local gradients need to be shared among workers at every iteration, SSGD performance is significantly influenced by network bottlenecks caused by either heterogeneous environment or bandwidth contention. To solve this problem, asynchronous parallel (ASP) strategy allows each worker to update parameters independently without synchronization, while suffering from accuracy loss and convergence inefficiency. In this paper, we propose a novel similarity-based synchronization scheme called SIM, which mitigates the impact of network bottlenecks and ensures convergence efficiency. Specifically, SIM reduces the number of aggregation workers based on the gradient similarity between global and local gradients, therefore shrinking the waiting time for the stragglers. We provide a theoretical analysis of convergence efficiency and conduct large-scale testbed experiments on CIFAR-10 and SQUAD dataset. The experimental results show that SIM reduces the convergence time of four classical deep learning models by up to 40%.
- Open Access
- 6 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 73-79
ORC: Online Reinforcement Learning for Congestion Control with Fast Convergence
- 6 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 73-79
Open AccessORC: Online Reinforcement Learning for Congestion Control with Fast Convergence
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1612-1622
SwitchTop- k : Scaling Top- k Compression on Programmable Switches
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1612-1622
Open AccessSwitchTop- k : Scaling Top- k Compression on Programmable Switches
- 27 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 33 (6) , 2897-2912
Progress-Aware Transmission Protocol for Efficient In-Network Aggregation in Distributed Machine Learning
Large-scale machine learning typically adopts distributed machine learning (DML) techniques to accelerate model training. Due to the large communication overhead, unfortunately, the phase of gradient aggregation has become the performance bottleneck for data-parallel DML. To reduce traffic volume, several in-network aggregation (INA) transmission protocols are proposed to offload gradient aggregation function into the programmable switches. However, since existing INA transmission protocols use synchronous congestion control mechanism to drive each round of gradient aggregation, the straggling workers lead to long iteration time and significant performance degradation. Besides, we reveal that existing INA solutions cannot provide the fairness performance among multiple jobs with varying number of workers. To solve the above problem, we propose PA-ATP, a progress-aware INA transmission protocol, which adopts the progress-aware asynchronous congestion control. PA-ATP adjusts the sending rate in accordance with the transmission progress, allowing the straggling flow to grab more bandwidth than the leading flow and control the asynchronous degree of straggling job. Moreover, to ensure the fair throughput among multiple jobs, we dynamically adjust the aggregator allocation for each job by tuning the number of hash operations. We use a P4 programmable switch and a kernel-bypass protocol stack to implement PA-ATP. The results of testbed and large-scale NS3 simulations show that PA-ATP reduces training time by up to 62% compared to the state-of-the-art INA transmission protocols.
- 27 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 33 (6) , 2897-2912
Progress-Aware Transmission Protocol for Efficient In-Network Aggregation in Distributed Machine Learning
Large-scale machine learning typically adopts distributed machine learning (DML) techniques to accelerate model training. Due to the large communication overhead, unfortunately, the phase of gradient aggregation has become the performance bottleneck for data-parallel DML. To reduce traffic volume, several in-network aggregation (INA) transmission protocols are proposed to offload gradient aggregation function into the programmable switches. However, since existing INA transmission protocols use synchronous congestion control mechanism to drive each round of gradient aggregation, the straggling workers lead to long iteration time and significant performance degradation. Besides, we reveal that existing INA solutions cannot provide the fairness performance among multiple jobs with varying number of workers. To solve the above problem, we propose PA-ATP, a progress-aware INA transmission protocol, which adopts the progress-aware asynchronous congestion control. PA-ATP adjusts the sending rate in accordance with the transmission progress, allowing the straggling flow to grab more bandwidth than the leading flow and control the asynchronous degree of straggling job. Moreover, to ensure the fair throughput among multiple jobs, we dynamically adjust the aggregator allocation for each job by tuning the number of hash operations. We use a P4 programmable switch and a kernel-bypass protocol stack to implement PA-ATP. The results of testbed and large-scale NS3 simulations show that PA-ATP reduces training time by up to 62% compared to the state-of-the-art INA transmission protocols.
- 22 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (10) , 9774-9790
Proactive Transport With High Link Utilization Using Opportunistic Packets in Cloud Data Centers
To meet the stringent demanding low latency and high throughput of cloud datacenter applications, recent receiver-driven transport protocols transmit only one packet once receiving each credit packet from the receiver to achieve ultra-low queueing delay. However, the round-trip time variation and the highly dynamic background traffic significantly deteriorate the performance of receiver-driven transport protocols, resulting in under-utilized bandwidth. This paper designs a simple yet effective solution called RPO, which retains the advantages of receiver-driven transmission while efficiently utilizing the available bandwidth. Specifically, RPO rationally uses low-priority opportunistic packets to ensure high network utilization without increasing the queueing delay of high-priority normal packets. Furthermore, to tackle the queueing buildup due to line-rate transmission in the first RTT, we design a selective dropping mechanism called SDM to help the majority of small flows complete within only one RTT by prioritizing the first-RTT bursty packets over the packets triggered by grants. We implement RPO in Linux hosts with DPDK. The experimental results show that RPO significantly improves the network utilization by up to 35% over the state-of-the-art schemes, without introducing additional queueing delay. Moreover, RPO integrated with SDM reduces the AFCT of small flows by up to 45% compared with RPO integrated with Aeolus.
- 22 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (10) , 9774-9790
Proactive Transport With High Link Utilization Using Opportunistic Packets in Cloud Data Centers
To meet the stringent demanding low latency and high throughput of cloud datacenter applications, recent receiver-driven transport protocols transmit only one packet once receiving each credit packet from the receiver to achieve ultra-low queueing delay. However, the round-trip time variation and the highly dynamic background traffic significantly deteriorate the performance of receiver-driven transport protocols, resulting in under-utilized bandwidth. This paper designs a simple yet effective solution called RPO, which retains the advantages of receiver-driven transmission while efficiently utilizing the available bandwidth. Specifically, RPO rationally uses low-priority opportunistic packets to ensure high network utilization without increasing the queueing delay of high-priority normal packets. Furthermore, to tackle the queueing buildup due to line-rate transmission in the first RTT, we design a selective dropping mechanism called SDM to help the majority of small flows complete within only one RTT by prioritizing the first-RTT bursty packets over the packets triggered by grants. We implement RPO in Linux hosts with DPDK. The experimental results show that RPO significantly improves the network utilization by up to 35% over the state-of-the-art schemes, without introducing additional queueing delay. Moreover, RPO integrated with SDM reduces the AFCT of small flows by up to 45% compared with RPO integrated with Aeolus.
- 6 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 74 (4) , 1362-1376
Asynchronous Control Based Aggregation Transport Protocol for Distributed Deep Learning
With the rapid growth scale of dataset and model, the training of deep neural networks (DNN) tends to be deployed in a distributed manner. In the large-scale distributed training, the bottlenecks have gradually moved from computational resources to communication process. Recent researches adopt in-network aggregation (INA) that offloads the gradient aggregation process to programmable switches, thereby reducing network traffic amount and transmission latency. Unfortunately, due to the bandwidth competition in shared training clusters, the straggler will slow down the training efficiency of INA. To address this issue, we propose an Asynchronous Control based Aggregation Transport Protocol (AC-ATP), which makes full use uncongested links to transmit gradients and the switch memory to cache gradients from the fast workers to accelerate the gradient aggregation. Meanwhile, AC-ATP performs congestion control according to the transmission progress of worker and the remaining completion time of the job. The evaluation results of real testbed and large-scale simulations show that AC-ATP reduces the aggregate time by up to 68% and speeds up training in real-world benchmark models.
- 6 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 74 (4) , 1362-1376
Asynchronous Control Based Aggregation Transport Protocol for Distributed Deep Learning
With the rapid growth scale of dataset and model, the training of deep neural networks (DNN) tends to be deployed in a distributed manner. In the large-scale distributed training, the bottlenecks have gradually moved from computational resources to communication process. Recent researches adopt in-network aggregation (INA) that offloads the gradient aggregation process to programmable switches, thereby reducing network traffic amount and transmission latency. Unfortunately, due to the bandwidth competition in shared training clusters, the straggler will slow down the training efficiency of INA. To address this issue, we propose an Asynchronous Control based Aggregation Transport Protocol (AC-ATP), which makes full use uncongested links to transmit gradients and the switch memory to cache gradients from the fast workers to accelerate the gradient aggregation. Meanwhile, AC-ATP performs congestion control according to the transmission progress of worker and the remaining completion time of the job. The evaluation results of real testbed and large-scale simulations show that AC-ATP reduces the aggregate time by up to 68% and speeds up training in real-world benchmark models.
- 19 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 32 (6) , 4917-4930
Straggler-Aware Gradient Aggregation for Large-Scale Distributed Deep Learning System
Deep Neural Network (DNN) is a critical component of a wide range of applications. However, with the rapid growth of the training dataset and model size, communication becomes the bottleneck, resulting in low utilization of computing resources. To accelerate communication, recent works propose to aggregate gradients from multiple workers in the programmable switch to reduce the volume of exchanged data. Unfortunately, since using synchronization transmission to aggregate data, current in-network aggregation designs suffer from the straggler problem, which often occurs in shared clusters due to resource contention. To address this issue, we propose a straggler-aware aggregation transport protocol (SA-ATP), which enables the leading worker to leverage the spare computing and storage resources to help the straggling worker. We implement SA-ATP atop clusters using P4-programmable switches. The evaluation results show that SA-ATP reduces the iteration time by up to 57% and accelerates training by up to 1.8 $\times$ in real-world benchmark models.
- 19 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 32 (6) , 4917-4930
Straggler-Aware Gradient Aggregation for Large-Scale Distributed Deep Learning System
Deep Neural Network (DNN) is a critical component of a wide range of applications. However, with the rapid growth of the training dataset and model size, communication becomes the bottleneck, resulting in low utilization of computing resources. To accelerate communication, recent works propose to aggregate gradients from multiple workers in the programmable switch to reduce the volume of exchanged data. Unfortunately, since using synchronization transmission to aggregate data, current in-network aggregation designs suffer from the straggler problem, which often occurs in shared clusters due to resource contention. To address this issue, we propose a straggler-aware aggregation transport protocol (SA-ATP), which enables the leading worker to leverage the spare computing and storage resources to help the straggling worker. We implement SA-ATP atop clusters using P4-programmable switches. The evaluation results show that SA-ATP reduces the iteration time by up to 57% and accelerates training by up to 1.8 $\times$ in real-world benchmark models.