arXiv subjects

Jiamin Jiang

Publications and source records attributed to Jiamin Jiang.

12 recordsLinked to original sources

arXiv2026

KRCA: An Efficient Root Cause Analysis System in Hyper-scale Microservice Systems via Agentic AI

Hyper-scale microservice systems have become the standard infrastructure for large-scale Internet companies. These systems consist of numerous loosely coupled microservices that evolve independently through continuous development and deployment. Such complexity makes failures unavoidable, necessitating efficient Root Cause Analysis (RCA) to help Site Reliability Engineers (SREs) quickly localize root cause services and classify failure types. However, existing RCA methods often struggle to adapt to the extreme dynamism and massive scale of these systems. In this paper, we present KRCA, an end-to-end RCA system designed for hyper-scale microservice systems. To manage the vast search space, KRCA employs a multi-stage pipeline that begins with an API-level drilldown to isolate suspicious services. It then instantiates a skeleton-based causal graph from anomalous metrics to serve as a high-recall structural prior, before utilizing a memory-augmented multi-agent framework to verify causality and generate the final failure report. By combining structured causal constraints with multi-agent reasoning, KRCA employs balances diagnostic accuracy with the efficiency requirements of real-time production use. Experimental results show that KRCA achieves AC@1 scores of 0.88 and 0.79 for root cause service localization and failure type classification, outperforming the strongest baseline by at lease 31% in absolute gains. KRCA has been deployed in Kuaishou's production environment for over six months, reducing the average diagnosis time by 77.3%.

cs.SE↗

arXiv2026

MeshTok: Efficient Multi-Scale Tokenization for Scalable PDE Transformers

Conventional patchified Transformers operate on uniform spatial partitions, distributing computational effort evenly across the domain irrespective of local features. This inflexible tokenization scheme is inherently limited in its ability to efficiently represent and process solutions to complex PDEs. To address this, we propose MeshTok, an adaptive mesh refinement (AMR)-inspired tokenization and sequence modeling framework. This method selectively refines spatial regions exhibiting sharp gradients, transient features, or multiscale structures, generating a heterogeneous set of multiscale tokens defined on a fixed simulation grid. These tokens are processed within a unified Transformer sequence, enabling the model to simultaneously capture coarse-grained global context and fine-grained local details without requiring specialized architectural components. Although adaptive refinement moderately increases token count, it promotes a more targeted allocation of computational resources to physically informative regions, which we view as a practical inductive bias rather than a formal optimality guarantee. Experimental evaluations across multiple PDE families and benchmark datasets demonstrate that MeshTok consistently improves the efficiency-accuracy trade-off compared to uniform-grid baselines. This suggests adaptive multiscale tokenization as a scalable and generalizable design principle for neural PDE modeling. Code is available at https://github.com/SCAILab-USTC/MeshTok.

cs.LG↗

arXiv2026

Graph of States: Solving Abductive Tasks with Large Language Models

Logical reasoning encompasses deduction, induction, and abduction. However, while Large Language Models (LLMs) have effectively mastered the former two, abductive reasoning remains significantly underexplored. Existing frameworks, predominantly designed for static deductive tasks, fail to generalize to abductive reasoning due to unstructured state representation and lack of explicit state control. Consequently, they are inevitably prone to Evidence Fabrication, Context Drift, Failed Backtracking, and Early Stopping. To bridge this gap, we introduce Graph of States (GoS), a general-purpose neuro-symbolic framework tailored for abductive tasks. GoS grounds multi-agent collaboration in a structured belief states, utilizing a causal graph to explicitly encode logical dependencies and a state machine to govern the valid transitions of the reasoning process. By dynamically aligning the reasoning focus with these symbolic constraints, our approach transforms aimless, unconstrained exploration into a convergent, directed search. Extensive evaluations on two real-world datasets demonstrate that GoS significantly outperforms all baselines, providing a robust solution for complex abductive tasks. Code repo and all prompts: https://github.com/gaorch85/Graph-of-States.

cs.AI↗

arXiv2026

OpsAgent: An Evolving Multi-agent System for Incident Management in Microservices

Incident management (IM) is central to the reliability of large-scale microservice systems. Yet manual IM, where on-call engineers examine metrics, logs, and traces is labor-intensive and error-prone in the face of massive and heterogeneous observability data. Existing automated IM approaches often struggle to generalize across systems, provide limited interpretability, and incur high deployment costs, which hinders adoption in practice. In this paper, we present OpsAgent, a lightweight, self-evolving multi-agent system for IM that employs a training-free data processor to convert heterogeneous observability data into structured textual descriptions, along with a multi-agent collaboration framework that makes diagnostic inference transparent and auditable. To support continual capability growth, OpsAgent also introduces a dual self-evolution mechanism that integrates internal model updates with external experience accumulation, thereby closing the deployment loop. Comprehensive experiments on the OPENRCA benchmark demonstrate state-of-the-art performance and show that OpsAgent is generalizable, interpretable, cost-efficient, and self-evolving, making it a practically deployable and sustainable solution for long-term operation in real-world microservice systems. Notably, its deployment in Lenovo's production environment further validates its effectiveness in real-world industrial settings.

cs.AI↗

arXiv2026

A Structure-Preserving Graph Neural Solver for Parametric Hyperbolic Conservation Laws

Hyperbolic conservation laws govern a wide range of transport-driven dynamics featuring shocks, contact discontinuities, and complex wave interactions, posing distinct challenges for deep-learning-based surrogate modeling. While classical numerical methods provide robust and physically admissible solutions, their computational cost restricts applicability in many-query tasks such as parametric studies and design optimization. Conversely, existing neural surrogates offer rapid inference but often fail to respect intrinsic PDE structures, leading to non-physical artifacts, rollout instability, and poor generalization. We present an interpretable, structure-preserving graph neural solver that bridges classical numerical principles with graph neural networks (GNNs). The network is designed as a learned reconstruction-and-flux operator rather than a black-box state updater, thereby inherently preserving key properties such as local conservation and upwinding. Inspired by Arbitrary high-order DERivatives schemes, we further recast message-passing GNNs as high-order space-time predictors, enabling conservative and stable neural updates with large time steps. Evaluation is performed on challenging supersonic flow benchmarks spanning broad parametric variations in geometry, initial/boundary conditions, and flow regimes. The neural solver achieves superior long-horizon rollout stability and accuracy compared with strong surrogate baselines, outperforms low-order discretizations, and delivers orders-of-magnitude runtime speedups over high-resolution simulations.

physics.comp-ph↗

arXiv2024

A Multigrid Graph U-Net Framework for Simulating Multiphase Flow in Heterogeneous Porous Media

Numerical simulation of multi-phase fluid dynamics in porous media is critical to a variety of geoscience applications. Data-driven surrogate models using Convolutional Neural Networks (CNNs) have shown promise but are constrained to regular Cartesian grids and struggle with unstructured meshes necessary for accurately modeling complex geological features in subsurface simulations. To tackle this difficulty, we build surrogate models based on Graph Neural Networks (GNNs) to approximate space-time solutions of multi-phase flow and transport processes. Particularly, a novel Graph U-Net framework, referred to as AMG-GU, is developed to enable hierarchical graph learning for the parabolic pressure component of the coupled partial differential equation (PDE) system. Drawing inspiration from aggregation-type Algebraic Multigrid (AMG), we propose a graph coarsening strategy adapted to heterogeneous PDE coefficients, achieving an effective graph pooling operation. Results of three-dimensional heterogeneous test cases demonstrate that the multi-level surrogates predict pressure and saturation dynamics with high accuracy, significantly outperforming the single-level baseline. Our Graph U-Net model exhibits great generalization capability to unseen model configurations.

physics.comp-ph↗

arXiv2024

Graph Convolutional Networks for Simulating Multi-phase Flow and Transport in Porous Media

Numerical simulation of multi-phase fluid dynamics in porous media is critical for many energy and environmental applications in Earth's subsurface. Data-driven surrogate modeling provides computationally inexpensive alternatives to high-fidelity numerical simulators. While the commonly used convolutional neural networks (CNNs) are powerful in approximating partial differential equation solutions, it remains challenging for CNNs to handle irregular and unstructured simulation meshes. However, simulation models for Earth's subsurface often involve unstructured meshes with complex mesh geometries, which limits the application of CNNs. To address this challenge, we construct surrogate models based on Graph Convolutional Networks (GCNs) to approximate the spatial-temporal solutions of multi-phase flow and transport processes in porous media. We propose a new GCN architecture suited to the hyperbolic character of the coupled PDE system, to better capture transport dynamics. Results of 2D heterogeneous test cases show that our surrogates predict the evolutions of pressure and saturation states with high accuracy, and the predicted rollouts remain stable for multiple timesteps. Moreover, the GCN-based models generalize well to irregular domain geometries and unstructured meshes that are unseen in the training dataset.

physics.comp-ph↗

arXiv2021

Inexact Methods for Sequential Fully Implicit (SFI) Reservoir Simulation

The sequential fully implicit (SFI) scheme was introduced (Jenny et al. 2006) for solving coupled flow and transport problems. Each time step for SFI consists of an outer loop, in which there are inner Newton loops to implicitly and sequentially solve the pressure and transport sub-problems. In standard SFI, the sub-problems are usually solved with tight tolerances at every outer iteration. This can result in wasted computations that contribute little progress towards the coupled solution. The issue is known as `over-solving'. Our objective is to minimize the cost of inner solvers while maintaining the convergence rate of SFI. We first extended a nonlinear-acceleration (NA) framework (Jiang and Tchelepi 2019) to multi-component compositional models, for ensuring robust outer-loop convergence. We then developed inexact-type methods that alleviate `over-solving'. It is found that there is no need for one sub-problem to strive for perfection, while the coupled (outer) residual remains high due to the other sub-problem. The new SFI solver was tested using several complex cases. The problems involve multi-phase and EoS-based compositional fluid systems. We compared different strategies such as fixed relaxations on absolute and relative tolerances for the inner solvers, as well as an adaptive approach. The results show that the basic SFI method is quite inefficient. Away from a coupled solution, additional accuracy achieved in inner solvers is wasted, contributing to little or no reduction of the overall outer residual. By comparison, the adaptive inexact method provides relative tolerances adequate for the current convergence state of the sub-problems. We show across a wide range of flow conditions that the new solver can effectively resolve the over-solving issue, and thus greatly improve the overall efficiency.

math.NA↗

arXiv2020

Localized Nonlinear Solution Strategies for Efficient Simulation of Unconventional Reservoirs

Accurate and efficient numerical simulation of unconventional reservoirs is challenging. Long periods of transient flow and steep potential gradients occur due to the extreme conductivity contrast between matrix and fracture. Detailed near-well/near-fracture models are necessary to provide sufficient resolution, but they are computationally impractical for field cases with multiple fracture stages. Previous works in the literature of unconventional simulations mainly focus on gridding level that adapts to wells and fractures. Limited research has been conducted on nonlinear strategies that exploit locality across timesteps and nonlinear iterations. To perform localized computations, an a-priori strategy is essential to first determine the active subset of simulation cells for the subsequent iteration. The active set flags the cells that will be updated, and then the corresponding localized linear system is solved. This work develops localization methods that are readily applicable to complex fracture networks and flow physics in unconventional reservoirs. By utilizing the diffusive nature of pressure updates, an adaptive algorithm is proposed to make adequate estimates for the active domains. In addition, we develop a localized solver based on nonlinear domain decomposition (DD). Comparing to a standard DD method, domain partitions are dynamically constructed. The new solver provides effective partitioning that adapts to flow dynamics and Newton updates. We evaluate the developed methods using several complex problems with discrete fracture networks. The results show that large degrees of solution locality present across timesteps and iterations. Comparing to a standard Newton solver, the new solvers enable superior computational performance. Moreover, Newton convergence behavior is preserved, without any impact on solution accuracy.

cs.CE↗

arXiv2020

Smooth Formulation for Isothermal Compositional Simulation with Improved Nonlinear Convergence

Compositional simulation is challenging, because of highly nonlinear couplings between multi-component flow in porous media with thermodynamic phase behavior. The coupled nonlinear system is commonly solved by the fully-implicit scheme. Various compositional formulations have been proposed. However, severe convergence issues of Newton solvers can arise under the conventional formulations. Crossing phase boundaries produces kinks in discretized equations, and subsequently causing oscillations or even divergence of Newton iterations. The objective of this work is to develop a smooth formulation that removes all the property switches and discontinuities associated with phase changes. We show that it can be very difficult and costly to smooth the conservation equations directly. Therefore, we first reformulate the coupled system, so that the discontinuities are transferred to the phase equilibrium model. In this way a single and concise non-smooth equation is achieved and then a smoothing approximation can be made. The new formulation with a smoothing parameter provides smooth transitions of variables across all the phase regimes. In addition, we employ a continuation method where the solution progressively evolves toward the target system. We evaluate the efficiency of the new smooth formulation and the continuation method using several complex problems. Compared to the standard natural formulation, the developed formulation and method exhibit superior nonlinear convergence behaviors. The continuation method leads to smooth and stable iterative performance, with a negligible impact on solution accuracy. Moreover, it works robustly for a wide range of flow conditions without parameter tuning.

physics.comp-ph↗

arXiv2020

Nonlinear Acceleration of Sequential Fully Implicit (SFI) Method for Coupled Flow and Transport in Porous Media

The sequential fully implicit (SFI) method was introduced along with the development of the multiscale finite volume (MSFV) framework, and has received considerable attention in recent years. Each time step for SFI consists of an outer loop to solve the coupled system, in which there is one inner Newton loop to implicitly solve the pressure equation and another loop to implicitly solve the transport equations. Limited research has been conducted that deals with the outer coupling level to investigate the convergence performance. In this paper we extend the basic SFI method with several nonlinear acceleration techniques for improving the outer-loop convergence. Specifically, we consider numerical relaxation, quasi-Newton (QN) and Anderson acceleration (AA) methods. The acceleration techniques are adapted and studied for the first time within the context of SFI for coupled flow and transport in porous media. We reveal that the iterative form of SFI is equivalent to a nonlinear block Gauss-Seidel (BGS) process. The effectiveness of the acceleration techniques is demonstrated using several challenging examples. The results show that the basic SFI method is quite inefficient, suffering from slow convergence or even convergence failure. In order to better understand the behaviors of SFI, we carry out detailed analysis on the coupling mechanisms between the sub-problems. Compared with the basic SFI method, superior convergence performance is achieved by the acceleration techniques, which can resolve the convergence difficulties associated with various types of coupling effects. We show across a wide range of flow conditions that the acceleration techniques can stabilize the iterative process, and largely reduce the outer iteration count.

physics.comp-ph↗

arXiv2020

Dissipation-Based Continuation Method for Multiphase Flow in Heterogeneous Porous Media

In reservoir simulation, solution of the coupled systems of nonlinear algebraic equations that are associated with fully-implicit (backward Euler) discretization is challenging. Having a robust and efficient nonlinear solver is necessary in order for reservoir simulation to serve as the primary tool for managing the recovery processes of large-scale reservoirs. Here, we develop a continuation method based on the use of a dissipation operator. We focus on nonlinear two-phase flow and transport in heterogeneous formations in the presence of viscous, gravitational, and capillary forces. The homotopy is constructed by adding numerical dissipation to the coupled discrete conservation equations. A continuation parameter is introduced to control the amount of dissipation. Numerical evidence of multi-dimensional models and detailed analysis of single-cell problems are used to explain how the dissipation operator improves the nonlinear convergence of the coupled system of equations. An adaptive strategy to determine the dissipation coefficient is proposed. The dissipation level is computed locally for each cell interface. We demonstrate the efficiency of the dissipation-based continuation (DBC) nonlinear solver using several examples, including 1D scalar transport and 2D heterogeneous problems with fully-coupled flow and transport. The DBC solver has better convergence properties compared with the standard damped-Newton solvers used in reservoir simulation.

physics.comp-ph↗