Xiaorui Wu

Xiaorui Wu

Auto-generated profile, based on publicly available information

Xiaorui Wu

Electric Power Research Institute of Guangxi Power Grid Co., Ltd,Southern Power Grid Corporation Wireless Power Transmission Joint Laboratory,Nanning,China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 June 2019 in IEEE/ACM Transactions on Networking
Verified
Journal Article
Closed access
GTGeorge Trimponias
George Trimponias
YXYan Xiao
Yan Xiao
XWXiaorui Wu
Xiaorui Wu
HXHong Xu
Hong Xu
YGYanhui Geng
Yanhui Geng
Traffic engineering (TE) is a fundamental task in networking. Conventionally, traffic can take any path connecting the source and destination. Emerging technologies such as segment routing, however, use logical paths that are composed of shortest paths going through a predetermined set of middlepoints in order to reduce the flow table overhead of TE implementation. Inspired by this, in this paper, we introduce the problem of node-constrained TE, where the traffic must go through a set of middlepoints, and study its theoretical fundamentals. We show that the general node-constrained TE that allows the traffic to take any path going through one or more middlepoints is NP-hard for directed graphs but strongly polynomial for undirected graphs, unveiling a profound dichotomy between the two cases. We also investigate a variant of node-constrained TE that uses only shortest paths between middlepoints, and prove that the problem can now be solved in weakly polynomial time for a fixed number of middlepoints, which explains why existing work focuses on this variant. Yet, if we constrain the end-to-end paths to be acyclic, the problem can become NP-hard. An important application of our work concerns flow centrality, for which we are able to derive complexity results. Furthermore, we investigate the middlepoint selection problem in general node-constrained TE. We introduce and study group flow centrality as a solution concept, and show that it is monotone but not submodular. Our work provides a thorough theoretical treatment of node-constrained TE and sheds light on the development of the emerging node-constrained TE in practice.

3 August 2020
Verified
Conference Paper
Open access
XWXiaorui Wu
Xiaorui Wu
HXHong Xu
Hong Xu
YWYi Wang
Yi Wang
DNN inference is becoming prevalent for many real-world applications. Current machine learning frameworks usually schedule inference tasks with the goal of optimizing throughput under predictable workloads and task arrival patterns. Yet, inference workloads are becoming more dynamic with bursty queries generated by various video analytics pipelines which run expensive inference only on a fraction of video frames. Thus it is imperative to optimize the completion time of these unpredictable queries and improve customer experience. We propose the preliminary design of the first online inference task scheduling system, called Irina, that takes completion time under unpredictable workload as its primary objective. Irina augments the design space of inference task scheduling with three new strategies, namely batching, stacking, and preemption, in order to more flexibly schedule the tasks and reduce overall latency. Simulation results with empirical inference execution data shows that Irina can improve average task completion time by 1.3x–2.5x over TensorFlow Serving scheduling.

6 April 2020 in IEEE Transactions on Services Computing
Verified
Journal Article
Closed access
XWXiaorui Wu
Xiaorui Wu
HXHong Xu
Hong Xu
BLBo Li
Bo Li
YXYongqiang Xiong
Yongqiang Xiong
The parameter server architecture is prevalently used for distributed deep learning. Each worker machine in a such system trains the complete model, which leads to a large amount of network data transfer between workers and servers. We empirically observe that the data transfer has a major impact on training time. We present a new distributed training system called Stanza to tackle this problem. Stanza exploits the fact that in many models such as convolution neural networks, most data exchange is attributed to the fully connected layers, while most computation is carried out in convolutional layers. Thus, we propose layer separation in distributed training: most nodes of the cluster train only the convolutional layers, while the rest train the fully connected layers. Gradients and parameters of the fully connected layers no longer need to be exchanged across the entire cluster, thereby substantially reducing the data transfer volume. We implement Stanza on PyTorch and evaluate its performance on Azure and EC2. Results show that Stanza accelerates training significantly over current parameter server systems: on EC2 instances with Tesla V100 GPU and 10Gb bandwidth for example, Stanza is 1.34x–13.9x faster for common deep learning models.

27 October 2023
Verified
Conference Paper
Closed access
XWXiaorui Wu
Xiaorui Wu
YMYuhong Mo
Yuhong Mo
JWJingyue Wang
Jingyue Wang
SCShaonan Chen
Shaonan Chen
JXJing Xiao
Jing Xiao
WGWenlan Gong
Wenlan Gong
There will be eddy current effect while foreign bodies inevitably enter the gap between the transmitting coil and the receiving coil in the process of magnetic field induction radio energy transmission, which could lead to the high temperature and inefficiency of the system or even significant security risks. In this paper, a detection method of metal foreign bodies in the auxiliary coil perpendicular to the energy coil is designed. The magnetic coupling mechanism model based on MAXWELL software was built to obtain parameters of the magnetic coupling mechanism by simulation. On this basis, the Simulink circuit simulation model of radio energy transmission system is established, and its energy transmission characteristics are analyzed. Based on the designed radio energy transmission system, a detection coil perpendicular to the energy coil was designed, and its MAXWELL 3D simulation model was established to obtain its coupling characteristics. The experimental prototype is built and the metal foreign body detection experiment is conducted. The results of experiment are consistent with that of the simulation, which verifies that the vertical detection coil proposed in this paper is a feasible method for the detection of metal foreign bodies in the radio energy transmission system.

27 October 2023
Verified
Conference Paper
Closed access
XWXiaorui Wu
Xiaorui Wu
YMYuhong Mo
Yuhong Mo
FMFanbin Meng
Fanbin Meng
Ning WuNing Wu
Ning Wu
JXJing Xiao
Jing Xiao
SCShaonan Chen
Shaonan Chen
IPT system is widely used in wireless charging of EVs. In this paper, a simulation model of typical IPT system based on LCC-LCC resonant network is established. By conducting simulations and analyzing the energy and efficiency traits of the system when exposed to varying load resistance, adjustable coupling coefficients, and frequency fluctuations, we obtain energy and efficiency curves for the system across different operational scenarios, which is of reference significance for the design of IPT system based on LCC-LCC resonant network.

26 August 2023
Verified
Conference Paper
Closed access
JXJing Xiao
Jing Xiao
HSHaonan Shi
Haonan Shi
SCShaonan Chen
Shaonan Chen
XWXiaorui Wu
Xiaorui Wu
WGWenlan Gong
Wenlan Gong
YMYuhong Mo
Yuhong Mo
For electric vehicles wireless power transfer (EV-WPT), the existing foreign object detection (FOD) solutions all use a switch mode power supply to power the FOD signal board. However, the transmit coil itself generates a magnetic field, which also has energy. Using this energy to replace the Switch Mode Power Supply can simplify the system structure, reduce energy waste, and integrate the signal board with the transmitting coil module to avoid significant interference caused by long communication lines. This article proposes a scheme to modify the structure of the transmitting coil, so that the externally assisted self-powering coil can utilize the energy of the transmitting coil while reducing interference with the foreign object detection coil array and receiving coil, achieving non-interference self-powering supply to the detection circuit.