XSXiangzhuang Shen

Xiangzhuang Shen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xiangzhuang Shen has a total of 129 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    HEAL: Performance Troubleshooting Deep inside Data Center Hosts

    by Yicheng Pan, Yang Zhang, Tingzhu Bi, Linlin Han, Yu Zhang, Meng Ma, Xiangzhuang Shen, Xinrui Jiang, Feng Wang, Xian Liu, et al.

    This study demonstrates the salient facts and challenges of host failure operations in hyperscale data centers. A host incident can involve hundreds of distinct host-level metrics. The faulting mechanism inside the host connects these heterogeneous metrics through direct and indirect correlation, making it extremely difficult to sort out the propagation procedures and the root cause from these intertwined indicators. To deeply understand the failure mechanism inside the host, we develop HEAL -- a novel host metrics analysis toolkit. HEAL discovers dynamic causality in sparse heterogeneous host metrics by combining the strengths of both time series and random variable analysis. It also extracts causal directional hints from causality's asymmetry and historical knowledge, which finally help HEAL produce accurate results given undesirable inputs. Evaluations in our production environment verify that HEAL provides significantly better result accuracy and full-process interpretability than the SOTA baselines. With these advantages, HEAL successfully serves our data center and worldwide product operations and impressively contributes to many other workflows.

  • Open Access

    HEAL: Performance Troubleshooting Deep inside Data Center Hosts

    by Yicheng Pan, Yang Zhang, Tingzhu Bi, Linlin Han, Yu Zhang, Meng Ma, Xiangzhuang Shen, Xinrui Jiang, Feng Wang, Xian Liu, et al.

    This study demonstrates the salient facts and challenges of host failure operations in hyperscale data centers. A host incident can involve hundreds of distinct host-level metrics. The faulting mechanism inside the host connects these heterogeneous metrics through direct and indirect correlation, making it extremely difficult to sort out the propagation procedures and the root cause from these intertwined indicators. To deeply understand the failure mechanism inside the host, we develop HEAL -- a novel host metrics analysis toolkit. HEAL discovers dynamic causality in sparse heterogeneous host metrics by combining the strengths of both time series and random variable analysis. It also extracts causal directional hints from causality's asymmetry and historical knowledge, which finally help HEAL produce accurate results given undesirable inputs. Evaluations in our production environment verify that HEAL provides significantly better result accuracy and full-process interpretability than the SOTA baselines. With these advantages, HEAL successfully serves our data center and worldwide product operations and impressively contributes to many other workflows.

  • Open Access

    HEAL: Performance Troubleshooting Deep inside Data Center Hosts

    by Yicheng Pan, Yang Zhang, Tingzhu Bi, Linlin Han, Yu Zhang, Meng Ma, Xiangzhuang Shen, Xinrui Jiang, Feng Wang, Xian Liu, et al.

    This study demonstrates the salient facts and challenges of host failure operations in hyperscale data centers. A host incident can involve hundreds of distinct host-level metrics, covering broad aspects. The faulting mechanism inside the host connects these heterogeneous metrics through direct and indirect correlation, making it extremely difficult to sort out the propagation procedures and the root cause from these intertwined indicators. To deeply understand the failure mechanism inside the host, we develop HEAL -- a novel host metrics analysis toolkit. HEAL synergistically discovers dynamic causality in sparse heterogeneous host metrics by combining the strengths of both time series and random variable analysis. It can also proactively extract causal directional hints from causality's asymmetry and historical knowledge. Together, these breakthroughs help HEAL produce accurate results given undesirable inputs. Extensive experiments in our production environment verify that HEAL provides significantly better result accuracy and full-process interpretability than the SOTA baselines. With these advantages, HEAL successfully serves our data center and worldwide product operations and impressively contributes to many other workflows.


  • Measurement of J/ψ decays into ΛbarLambda π + π −

    by M. Ablikim, J. Z. Bai, Y. Bai, Y. Ban, X. Cai, H. S. Chen, J. C. Chen, Y. B. Chen, Y. P. Chu, Y. S. Dai, et al.

    Based on 58 million J/ψ events collected by the BES II detector at the BEPC, J/ψ → Λ π+π− is observed for the first time. The branching fraction is measured to be Br(J/ψ → Λ π+π−) = (4.30±0.13±0.99)×10−3, excluding the decays to intermediate states, namely J/ψ → Ξ−+, J/ψ → Σ(1385)−(1385)+, and J/ψ → Σ(1385)+(1385)−. The branching fractions for these intermediate resonance channels are measured to be: Br(J/ψ → Ξ−+) = (0.90±0.03±0.18)×10−3, Br(J/ψ → Σ(1385)−(1385)+) = (1.23±0.07±0.30)×10−3, and Br(J/ψ → Σ(1385)+(1385)−) = (1.50±0.08±0.38)×10−3, respectively. The angular distribution is of the form ∝ (1 + αcos2θ) with α = (0.35±0.29±0.06) for J/ψ → Ξ−+, α = (−0.54±0.22±0.10) for J/ψ → Σ(1385)−(1385)+, and α = (−0.35±0.25±0.06) for J/ψ → Σ(1385)+(1385)−.