Detecting large-scale system problems by mining console logs
The DOI matches its Crossref record; 43 of the 44 references it cites were found in a registry or an index.
Abstract
Surprisingly, console logs rarely help operators detect problems in large-scale datacenter services, for they often consist of the voluminous intermixing of messages from many software components written by independent developers. We propose a general methodology to mine this rich source of information to automatically detect system runtime problems. We first parse console logs by combining source code analysis with information retrieval to create composite features. We then analyze these features using machine learning to detect operational problems. We show that our method enables analyses that are impossible with previous methods because of its superior ability to create sophisticated features. We also show how to distill the results of our analysis to an operator-friendly one-page decision tree showing the critical messages associated with the detected problems. We validate our approach using the Darkstar online game server and the Hadoop File System, where we detect numerous real problems with high accuracy and few false positives. In the Hadoop case, we are able to analyze 24 million lines of console logs in 3 minutes. Our methodology works on textual console logs of any size and requires no changes to the service software, no human input, and no knowledge of the software's internals.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Besides, Log2Vec also adopts MIMICK [49], an approach of inducing word embedding from character-level features to handle OOV words to improve anomaly detection performance.”
From Log-based Anomaly Detection Without Log Parsing (2021) · cites this paper as [49]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Jolliffe, I. (2005) · Encyclopedia of Statistics in Behavioral Science · doi:10.1002/0470013192.bsa501 · Cited by 14,310
Introduction to information retrieval
(2009) · Choice Reviews Online · doi:10.5860/choice.46-2715 · Cited by 12,278
Term-weighting approaches in automatic text retrieval
Salton, G., & Buckley, C. (1988) · Information Processing & Management · doi:10.1016/0306-4573(88)90021-0 · Cited by 9,740 · cited 2×
“We have investigated a variety of such methods and have found that Principal Component Analysis (PCA) [5, 16] combined with term-weighting techniques from information retrieval [23, 26] yields excellent anomaly detection results on both feature matrices, while requiring little parameter tuning.”
MethodWitten, I. H. (2008) · Cited by 9,143
Data mining: practical machine learning tools and techniques with Java implementations
Witten, I. H., & Frank, E. (2002) · ACM SIGMOD Record · doi:10.1145/507338.507355 · Cited by 5,137
The text mining handbook: advanced approaches in analyzing unstructured data
(2007) · Choice Reviews Online · doi:10.5860/choice.44-5684 · Cited by 1,598
Control Procedures for Residuals Associated With Principal Component Analysis
Jackson, J. E., & Mudholkar, G. S. (1979) · Technometrics · doi:10.1080/00401706.1979.10489779 · Cited by 1,237
Diagnosing network-wide traffic anomalies
Lakhina, A., Crovella, M., & Diot, C. (2004) · Proceedings of the 2004 conference on Applications, technologies, architectures, and protocols for computer communications · doi:10.1145/1015467.1015492 · Cited by 1,039
YALE: rapid prototyping for complex data mining tasks
Mierswa, I., Wurst, M., Klinkenberg, R., Scholz, M., & Euler, T. (2006) · Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1150402.1150531 · Cited by 1,010
Pinpoint: problem determination in large, dynamic Internet services
Chen, M. Y., Kiciman, E., Fratkin, E., Fox, A., & Brewer, E. (2003) · Proceedings International Conference on Dependable Systems and Networks · doi:10.1109/dsn.2002.1029005 · Cited by 835 · cited 1×
“, 2002; Ma & Hellerstein, 2001) to more detailed tracing such as execution paths (Chen et al., 2002; Chen & Brewer, 2004).”
BackgroundSelf-citation
Research works printed but not linked: 4
- [2] D. Borthakur. The hadoop distributed file system: Architecture and design. Hadoop Project Website, 2007. -
- [5] R. Dunia and S. J. Qin. Multi-dimensional fault diagnosis using a subspace approach. In Proc. ACC , 1997. -
- [8] R. Fonseca and et al. Xtrace: A pervasive network tracing framework. In In Proc. NSDI , 2007. -
- [14] W. Jiang and et al. Understanding customer problem troubleshooting from storage system logs. In Proceedings of USENIX FAST'09 , 2009. -
Cited by
Items 401–500 of 1,004
Self-supervised log parsing using semantic contribution difference
Yu, S., Chen, N., Wu, Y., & Dou, W. (2023) · Journal of Systems and Software · doi:10.1016/j.jss.2023.111646 · Cited by 22
MDFULog: Multi-Feature Deep Fusion of Unstable Log Anomaly Detection Model
Li, M., Sun, M., Li, G., Han, D., & Zhou, M. (2023) · Applied Sciences · doi:10.3390/app13042237 · Cited by 24
No, G., Lee, Y., Kang, H., & Kang, P. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.108613 · Cited by 24
Jin, H., Papadimitriou, G., Raghavan, K., Zuk, P., Balaprakash, P., Wang, C., Mandal, A., & Deelman, E. (2024) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1109/sc41406.2024.00098 · Cited by 26
“For example, [13] proposed a general methodology for mining console logs to detect large-scale system problems.”
Four Decades of Data Mining in Network and Systems Management
Kalegele, K., Sasai, K., Takahashi, H., Kitagata, G., & Kinoshita, T. (2015) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2015.2426713 · Cited by 21
“Detection [7], [23], [34], [35], [36], [39], [40], [42], [70], [74], [76], [77], [92] Analysis of NS-data for modeling (including correlation) the normal behaviors of a system so that some other desired behaviors (e.”
BackgroundNetAlytics: Cloud-Scale Application Performance Monitoring with SDN and NFV
Liu, G., Trotter, M., Ren, Y., & Wood, T. (2016) · International Middleware Conference · doi:10.1145/2988336.2988344 · Cited by 21
“Common approaches includes monitoring performance metrics [20, 38, 45], instrumenting applications [8, 14, 23, 42] and annotating logs [31, 41].”
BackgroundA Framework for Anomaly Detection in Time-Driven and Event-Driven Processes using Kernel Traces
Ezeme, O. M., Mahmoud, Q. H., & Azim, A. (2020) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2020.2978469 · Cited by 21
Automated evolution of feature logging statement levels using Git histories and degree of interest
Tang, Y., Spektor, A., Khatchadourian, R., & Bagherzadeh, M. (2021) · Science of Computer Programming · doi:10.1016/j.scico.2021.102724 · Cited by 21
“[60] mine log output to detect problems.”
Tiresias: Online Anomaly Detection for Hierarchical Operational Network Data
Hong, C.-Y., Caesar, M., Duffield, N. G., & Wang, J. (2012) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs.2012.30 · Cited by 21
“[26] also applied PCA to mine abnormal feature vectors in console logs.”
MethodLog Sequence Anomaly Detection Based on Template and Parameter Parsing via BERT
Chai, X., Zhang, H., Zhang, J., Sun, Y., & Das, S. K. (2024) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2024.3428538 · Cited by 26
“Among them, HDFS [39], Hadoop [40], Spark, ZooKeeper (ZK), and OpenStack (OS) are distributed system logs.”
Tan, Y., Liu, J., Li, Y., & Wang, J. (2025) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2025.3539660 · Cited by 20
“It was first exploited in [45] to detection anomalies based on log data.”
Proctor: Detecting and Investigating Interference in Shared Datacenters
Kannan, R. S., Jain, A., Laurenzano, M. A., Tang, L., & Mars, J. (2018) · IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) · doi:10.1109/ispass.2018.00016 · Cited by 20
“A Priori Knowledge Another class of applications observe correlation between various system parameters to detect performance issues in runtime and its root causes [1] [42] [46] [28] [24].”
BackgroundLogGrep: Fast and Cheap Cloud Log Storage by Exploiting both Static and Runtime Patterns
Wei, J., Zhang, G., Chen, J., Wang, Y., Zheng, W., Sun, T., Wu, J., & Jiang, J. (2023) · European Conference on Computer Systems · doi:10.1145/3552326.3567484 · Cited by 32
SSDLog: a semi-supervised dual branch model for log anomaly detection
Lu, S., Han, N., Wang, M., Wei, X., Lin, Z., & Wang, D. (2023) · World Wide Web · doi:10.1007/s11280-023-01174-y · Cited by 24
Smell-driven performance analysis for end-user programmers
Chambers, C., & Scaffidi, C. P. (2013) · IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC) · doi:10.1109/vlhcc.2013.6645261 · Cited by 20
Semantic Aware Online Detection of Resource Anomalies on the Cloud
Bhattacharyya, A., Jandaghi, S. A. J., Sotiriadis, S., & Amza, C. (2016) · IEEE International Conference on Cloud Computing Technology and Science · doi:10.1109/cloudcom.2016.0035 · Cited by 20
“By offering context based anomaly extraction with minimal processing cost, we anticipate that our solution will promote augmenting routine system monitoring with realtime statistical analysis for context-aware anomaly detection.”
MethodDlog: diagnosing router events with syslogs for anomaly detection
Li, T., Ma, J., & Sun, C. (2017) · The Journal of Supercomputing · doi:10.1007/s11227-017-2165-9 · Cited by 20
“Thus, we need to use the PCA algorithm [19] which was also used in many works [9,20] to deduce the dimension without losing the main features of the log chunks.”
BackgroundMethodInfluentialProfiling distributed systems in lightweight virtualized environments with logs and resource metrics
Pi, A., Chen, W., Zhou, X., & Ji, M. (2018) · ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC) · doi:10.1145/3208040.3208044 · Cited by 21
“In the future,we plan to use machine learning methods or rule-based methods to automatically build the relationship between logs and resource metrics, which further takes the burdens off users.”
MethodDILAF: A framework for distributed analysis of large‐scale system logs for anomaly detection
Astekin, M., Zengin, H., & Sözer, H. (2018) · Software Practice and Experience · doi:10.1002/spe.2653 · Cited by 23
“1 , 6 System logs are parsed to extract information from free text-based log messages.”
BackgroundMethodInfluentialLog-Based Anomaly Detection with Multi-Head Scaled Dot-Product Attention Mechanism
Du, Q., Zhao, L. Y., Xu, J., Han, Y., & Zhang, S. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-86472-9_31 · Cited by 23
Parkinson, S., & Khan, S. (2018) · Journal of Information Security and Applications · doi:10.1016/j.jisa.2018.03.003 · Cited by 20
“A similar study [19] has extracted identifiers and state variables via console logging of over 24 million messages and used Principal Component Analysis (PCA)- based unsupervised learning algorithm for anomaly detection.”
MethodEstimation of Dimensions Contributing to Detected Anomalies with Variational AutoencodersPreprint
Ikeda, Y., Tajiri, K., Nakano, Y., Watanabe, K., & Ishibashi, K. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.04576 · Cited by 20
“For the interpretation of anomalies, (Xu et al. 2009) proposed a combination algorithm of PCA and decision trees for visualizing how the anomalies are detected.”
MethodRetrieval-Grounded HDFS Log Anomaly Detection and Deterministic Failure Narrative Generation
Sun, X., Zhong, Z. S., & Wu, Q. (2026) · Journal of Computational Systems and Applications · doi:10.64229/j6d7fr94 · Cited by 20
A User and Entity Behavior Analytics Log Data Set for Anomaly Detection in Cloud Computing
Landauer, M., Skopik, F., Höld, G., & Wurzenberger, M. (2022) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata55660.2022.10020672 · Cited by 28
“Some of the most widely used data sets for log-based anomaly detection are the HDFS [17], BlueGene/L (BGL) [18], Thunderbird [18], OpenStack [19], Hadoop [20], and the data sets from Loghub [21].”
MethodUncertainty in runtime verification: A survey
Taleb, R., Hallé, S., & Khoury, R. (2023) · Computer Science Review · doi:10.1016/j.cosrev.2023.100594 · Cited by 26
Hu, Y., Zhang, D., Cao, G., & Pan, Q. (2019) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2019.8913895 · Cited by 19
“For example, for the ICS network data, the protocol types are [TCP, UDP, Modbus], then they can be quantified as [1, 2, 3].”
BackgroundMethodHybrid CAE-VAE for Unsupervised Anomaly Detection in Log File Systems
Wadekar, A., Gupta, T., Vijan, R., & Kazi, F. (2019) · International Conference on Computing, Communication and Networking Technologies (ICCCNT) · doi:10.1109/icccnt45670.2019.8944863 · Cited by 19
Incremental Analysis of Large-Scale System Logs for Anomaly Detection
Astekin, M., Ozcan, S., & Sözer, H. (2019) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata47090.2019.9006593 · Cited by 19
“Construct validity threats are mitigated by repeating the experiment with various samples of the benchmark dataset [1], [17], [18] that is commonly utilized in previous studies and well-known in the research domain.”
BackgroundMethodResultInfluentialMonitoring Anomalous States of Computer Systems by Intellectual Analysis of Data of System Journals
Sheluhin, O. I., Kostin, D. V., & Gorodnichev, M. G. (2020) · Systems of Signals Generating and Processing in the Field of on Board Communications · doi:10.1109/ieeeconf48371.2020.9078632 · Cited by 19
“The anomaly detection system consists of four successive stages: collecting initial data, analyzing and processing the logbook, extracting features and numerical representation, building an anomaly detection model [8, 12, 13, 14].”
MethodA Generative Adversarial Networks for Log Anomaly Detection
Duan, X., Shi, Y., Yuan, W., Cheng, H., & Yin, X. (2021) · Computer Systems Science and Engineering · doi:10.32604/csse.2021.014030 · Cited by 22
SiaLog: detecting anomalies in software execution logs using the siamese network
Hashemi, S., & Mäntylä, M. (2022) · Automated Software Engineering · doi:10.1007/s10515-022-00365-7 · Cited by 27
“One of the most well-known and effective anomaly detection methods is Principal Component Analysis (PCA), mentioned by Xu et al. (2009b).”
MethodLog Analytics for Dependable Enterprise Telephony
Chen, C., Singh, N., & Yajnik, S. (2012) · European Dependable Computing Conference · doi:10.1109/edcc.2012.14 · Cited by 19
“[14] uses source code analysis to identify important variables, groups messages based on these variables and applies machine-learning techniques to derive anomalies in the system.”
MethodFast entropy based alert detection in super computer logs
Makanju, A., Zincir-Heywood, N., & Milios, E. (2010) · International Conference on Dependable Systems and Networks workshops · doi:10.1109/dsnw.2010.5542621 · Cited by 19
A Convolutional Auto-Encoder Method for Anomaly Detection on System Logs
Cui, Y. J., Sun, Y., Hu, J., & Sheng, G. (2018) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2018.00519 · Cited by 19
Astekin, M., Zengin, H., & Sözer, H. (2018) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata.2018.8621967 · Cited by 19
“We utilized the manually labeled data to validate our results [2].”
BackgroundMethodResultInfluentialStatistical debugging for real-world performance problems
Song, L., & Lu, S. (2014) · ACM SIGPLAN Notices · doi:10.1145/2714064.2660234 · Cited by 19
“Many techniques been proposed to diagnose performance problems in distributed systems [2, 12, 20, 32, 39].”
BackgroundSummarizing Unstructured Logs in Online ServicesPreprint
Meng, W., Zaiter, F., Huang, Y., Liu, Y., Zhang, S., Zhang, Y., Zhu, Y., Zhang, T., Wang, E., Ren, Z., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.08938 · Cited by 24
“of ranked summaries on the gold references and compare it with the baselines. A. Experimental Setting 1) Datasets: We conduct experiments over four public log datasets from several services: BGL logs [28], HDFS logs [6], HPC logs [29], and Proxifier logs [3]. Since the lack of a publicly-available summary gold standard…”
MethodLogOnline: A Semi-Supervised Log-Based Anomaly Detector Aided with Online Learning Mechanism
Wang, X., Song, J., Zhang, X., Tang, J., Gao, W., & Lin, Q. (2023) · 2023 38th IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1109/ase56229.2023.00043 · Cited by 18
“These works can be roughly divided into two categories, i.e. statistical learning approaches [12], [14], [15], [36], [37], and deep learning methods [2]–[6], [38].”
BackgroundMethodInfluentialCross-Modal Attention Networks for Multi-Modal Anomaly Detection in System Software
Xing, S., & Wang, Y. (2025) · IEEE Open Journal of the Computer Society · doi:10.1109/ojcs.2025.3607975 · Cited by 88
“Early works relied on invariant mining [11] and Principal Component Analysis (PCA) [12] to identify patterns in structured log data.”
Multi-source data based anomaly detection through temporal and spatial characteristics
Xu, P., Gao, Q., Zhang, Z., & Zhao, K. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.121675 · Cited by 18
Log Parsing with Generalization Ability under New Log Types
Yu, S., Wu, Y., Li, Z., He, P., Chen, N., & Liu, C. (2023) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3611643.3616355 · Cited by 22
“Automated log parsing has replaced manual log reading and the manual setting of regular expressions through source code [46].”
MethodWei, X., Wang, J., Sun, C., Towey, D., Shoufeng, Z., Zuo, W., Yu, Y., Ruan, R., & Song, G. (2024) · Journal of Software Evolution and Process · doi:10.1002/smr.2650 · Cited by 26
“50 Parameter values in a log message often record important system state metrics.”
InfluentialMagnifier: Online Detection of Performance Problems in Large-Scale Cloud Computing Systems
Mi, H., Wang, H., Yin, G., Cai, H., Zhou, Q., Sun, T., & Zhou, Y. (2011) · IEEE International Conference on Services Computing · doi:10.1109/scc.2011.25 · Cited by 18
“There are mainly two kinds of solutions, black boxes based schema [1, 4] and instrumentation based schema [2, 3, 11, 14, 15, 19, 20, 22].”
BackgroundHow are distributed bugs diagnosed and fixed through system logs?
Wei, Y., Lu, S., Sun, H., & Liu, X. (2019) · Information and Software Technology · doi:10.1016/j.infsof.2019.106234 · Cited by 18
OILog: An online incremental log keyword extraction approach based on MDP-LSTM neural network
Duan, X., Ying, S., Cheng, H., Yuan, W., & Yin, X. (2020) · Information Systems · doi:10.1016/j.is.2020.101618 · Cited by 19
CloudSeer: Workflow Monitoring of Cloud Infrastructures via Interleaved Logs
Yu, X., Joshi, P., Xu, J., Jin, G., Zhang, H., & Jiang, G. (2016) · ACM SIGPLAN Notices · doi:10.1145/2954679.2872407 · Cited by 18
Detecting the performance impact of upgrades in large operational networks
Mahimkar, A. A., Song, H. H., Ge, Z., Shaikh, A., Wang, J., Yates, J. M., Zhang⋆, Y., & Emmons, J. (2010) · ACM SIGCOMM Computer Communication Review · doi:10.1145/1851275.1851219 · Cited by 18
P3CA: Private Anomaly Detection Across ISP Networks
Nagaraja, S., Jalaparti, V., Caesar, M., & Borisov, N. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-22263-4_3 · Cited by 18
“Extensions to PCA make it robust to attacks such as variance-injection [33], and enable PCA to be used for other goals, such as diagnosing network disruptions [17].”
MethodEnCore: exploiting system environment and correlation information for misconfiguration detection
Zhang, J., Renganarayana, L., Zhang, X., Ge, N., Bala, V., Xu, T., & Zhou, Y. (2014) · ACM SIGPLAN Notices · doi:10.1145/2644865.2541983 · Cited by 18
Similarity measurement’s comparison with mapping and localization in large-scale
Satea, H. D., Ibrahem, A. A., Faiq, M., Abbood, Z. A., & Ahmed, S. R. (2023) · AIP conference proceedings · doi:10.1063/5.0119964 · Cited by 17
Formal concept analysis approach to understand digital evidence relationships
Sokol, P., Antoni, Ľ., Krídlo, O., Marková, E., Kováčová, K., & Krajči, S. (2023) · International Journal of Approximate Reasoning · doi:10.1016/j.ijar.2023.108940 · Cited by 17
Graph neural networks based log anomaly detection and explanation
Li, Z., Shi, J., & van Leeuwen, M. (2026) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-026-01235-6 · Cited by 18
XDrain: Effective log parsing in log streams using fixed-depth forest
Liu, C., Tian, Y., Yu, S., Gao, D., Wu, Y., Huang, S., Hu, X., & Chen, N. (2024) · Information and Software Technology · doi:10.1016/j.infsof.2024.107546 · Cited by 18
LLM-Based Misconfiguration Detection for AWS Serverless Computing
Wen, J., Chen, Z., Zhu, Z., Sarro, F., Liu, Y., Ping, H., & Wang, S. (2026) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3745766 · Cited by 24
“To evaluate the effectiveness of approaches in identifying real-world misconfigurations in serverless applications, we construct a relevant dataset by mining real-world configuration issues from GitHub.”
Frequent Pattern Based User Behavior Anomaly Detection for Cloud System
Chiu, C.-Y., Yeh, C.-T., & Lee, Y. (2013) · Conference on Technologies and Applications of Artificial Intelligence (TAAI) · doi:10.1109/taai.2013.25 · Cited by 17
“Xu et al. [11] proposed a method to generate normal model of system output for detecting system problems from console logs.”
MethodH ansel: diagnosing faults in openStack
Sharma, D., Poddar, R., Mahajan, K., Dhawan, M., & Mann, V. (2015) · ACM International Conference on emerging Networking EXperiments and Technologies (CoNEXT) · doi:10.1145/2716281.2836108 · Cited by 17
“Some tracing systems [17, 30, 40, 42, 44] rely on log analysis to infer causal relationship among events.”
BackgroundPredictive Analysis in Network Function Virtualization
Li, Z., Ge, Z., Mahimkar, A. A., Wang, J., Zhao, B. Y., Zheng, H.-T., Emmons, J., & Ogden, L. (2018) · ACM Internet Measurement Conference (IMC) · doi:10.1145/3278532.3278547 · Cited by 17
“Syslogs are complex, unstructured, free-form texts generated by the systems to describe a wide range of events [26, 35].”
BackgroundMethodReal-Time Evasion Attacks against Deep Learning-Based Anomaly Detection from Distributed System Logs
Herath, J. D., Yang, P., & Yan, G. (2021) · ACM Conference on Data and Application Security and Privacy (CODASPY) · doi:10.1145/3422337.3447833 · Cited by 22
“Other unsupervised machine learning techniques include Statistical methods such as Principal Component Analysis (PCA) [42] that operate by reducing a high dimensional data space to a lower dimension, and techniques such as Invariant Mining [28] that identify common linear relationships in benign…”
BackgroundMethodInfluentialAnomalyAdapters: Parameter-Efficient Multi-Anomaly Task Detection
Ünal, U., & Dağ, H. (2022) · IEEE Access · doi:10.1109/access.2022.3141161 · Cited by 21
“two studies, PCA [9] which analyzes log representation as count vectors, DeepLog [12] which uses LSTM model to predict next log key in workflow.”
MethodInfluentialMi, H., Wang, H., Zhou, Y., Lyu, M. R.-T., & Cai, H. (2012) · Science China Information Sciences · doi:10.1007/s11432-012-4747-8 · Cited by 17
A Framework for System Event Classification and Prediction by Means of Machine Learning
Pitakrat, T., Grunert, J., Kabierschke, O., Keller, F., & van Hoorn, A. (2015) · International ICST Conference on Performance Evaluation Methodologies and Tools (VALUETOOLS) · doi:10.4108/icst.valuetools.2014.258197 · Cited by 17
“[12] detect problems in large-scale systems by combining source code analysis with machine learning to mine composite features and detect operational problems at runtime.”
BackgroundHound: Causal Learning for Datacenter-scale Straggler Diagnosis
Zheng, P., & Lee, B. C. (2018) · Proceedings of the ACM on Measurement and Analysis of Computing Systems · doi:10.1145/3179420 · Cited by 17
“Similarly, tree-based methods [56, 69] apply classification and regression trees to characterize the causal rules that produce specific performance outcomes.”
MADDC: Multi-Scale Anomaly Detection, Diagnosis and Correction for Discrete Event Logs
Wang, X., Yang, L., Li, D., Ma, L., He, Y., Xiao, J., Liu, J., & Yang, Y. (2022) · Proceedings of the 38th Annual Computer Security Applications Conference · doi:10.1145/3564625.3567972 · Cited by 16
“To answer these questions, we evaluate MADDC with three datasets fromUNSW-NB15 [31] intrusion detection traffic logs (simplified as UNSWNB) and HDFS system logs [38].”
BackgroundMethodInfluentialAnomaly detection in log-event sequences: A federated deep learning approach and open challenges
Himler, P., Landauer, M., Skopik, F., & Wurzenberger, M. (2024) · Machine Learning with Applications · doi:10.1016/j.mlwa.2024.100554 · Cited by 26
Diallo, M. H., August, M., Hallman, R. A., Kline, M., Slayback, S. M., & Graves, C. T. (2017) · Cluster Computing · doi:10.1007/s10586-017-0900-x · Cited by 16
“Xu et al. [31], designed and implemented a method of mining console logs to automatically detect system errors and anomalies.”
MethodEnsemble Methods for Anomaly Detection Based on System Log
Xia, X., Zhang, W., & Jiang, J. (2019) · IEEE Pacific Rim International Symposium on Dependable Computing (PRDC) · doi:10.1109/prdc47002.2019.00034 · Cited by 16
“We choose these models, which are LogCluster [4], Principal Component Analysis (PCA) [5] and Invariants mining [6] as our basic model then use assemble learning methods.”
MethodBoosted Ensemble Learning for Anomaly Detection in 5G RAN
Sundqvist, T., Bhuyan, M. H., Forsman, J., & Elmroth, E. (2020) · IFIP advances in information and communication technology · doi:10.1007/978-3-030-49161-1_2 · Cited by 23
“This data is described more in detail in [28] and has been used for research in anomaly detection by [25] and [9].”
MethodResultPerformance Evaluation of Anomaly Detection in Imbalanced System Log Data
Studiawan, H., & Sohel, F. A. (2020) · World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4) · doi:10.1109/worlds450073.2020.9210329 · Cited by 16
“The last two are unsupervised learning methods, including isolation forest [5] and principal component analysis (PCA) [6].”
MethodInfluentialOn Matching Log Analysis to Source Code: A Systematic Mapping Study
Bushong, V., Sanders, R., Curtis, J., Du, M., Cerny, T., Frajták, K., Bureš, M., Tisnovsky, P., & Shin, D. (2020) · ACM Conference on Research in Adaptive and Convergent Systems (RACS) · doi:10.1145/3400286.3418262 · Cited by 22
“These templates contain the static portion of the log message as well as the matching interpolated variable values [2, 5, 8, 9, 12, 14, 15, 22, 23, 25].”
BackgroundMethodInfluentialYuan, L.-P., Liu, P., & Zhu, S. (2021) · ACM Asia Conference on Computer and Communications Security (ASIA CCS) · doi:10.1145/3433210.3453098 · Cited by 18
“UNSW-NB15 [29] traffic logs and HDFS console logs [39].”
BackgroundMethodInfluentialLogAttn: Unsupervised Log Anomaly Detection with an AutoEncoder Based Attention Mechanism
Zhang, L., Li, W., Zhang, Z., Lu, Q., Hou, C., Hu, P. C., Gui, T., & Lu, S. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-82153-1_19 · Cited by 16
Chat-XAI: A New Chatbot to Explain Artificial Intelligence
Gao, M., Liu, X., Xu, A., & Akkiraju, R. (2021) · Lecture notes in networks and systems · doi:10.1007/978-3-030-82199-9_9 · Cited by 20
InterpretableSAD: Interpretable Anomaly Detection in Sequential Log Data
Han, X., Cheng, H., Xu, D., & Yuan, S. (2021) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata52589.2021.9671642 · Cited by 18
“• Hadoop Distributed File System (HDFS) [9].”
MethodInfluentialA Study of Failure Recovery and Logging of High-Performance Parallel File Systems
Han, R., Gatla, O. R., Zheng, M., Cao, J., Zhang, D., Dai, D., Chen, Y., & Cook, J. (2022) · ACM Transactions on Storage · doi:10.1145/3483447 · Cited by 21
“More sophisticated techniques in terms of log parsing and feature extraction [92] are likely needed to handle the PFS logs effectively.”
MethodPerformance problems diagnosis in cloud computing systems by mining request trace logs
Mi, H., Wang, H., Yin, G., Cai, H., Zhou, Q., & Sun, T. (2012) · IEEE/IFIP Network Operations and Management Symposium (NOMS) · doi:10.1109/noms.2012.6212005 · Cited by 16
“For each major cluster, we separate their requests into the normal set and the abnormal set according to the distances of their projection onto the abnormal space (we refer readers to the papers of [11], [12] for more details).”
BackgroundMethodBridging the divide between software developers and operators using logs
Shang, W. (2012) · doi:10.5555/2337223.2337490 · Cited by 16
Data Stream Clustering for Online Anomaly Detection in Cloud Applications
Sauvanaud, C., Silvestre, G., Kaâniche, M., & Kanoun, K. (2015) · European Dependable Computing Conference · doi:10.1109/edcc.2015.22 · Cited by 19
“[11], [25], [29], [33] are such examples, respectively working on detecting anomalies by means of Tree-Augmented Bayesian Networks (TAN), Principal Component Analysis (PCA) associated with decision trees, Semi-Hidden Markov models, and Two-dependent Markov models used conjointly with TAN.”
BackgroundMethodDelivering software with agility and quality in a cloud environment
Oliveira, F., Eilam, T., Nagpurkar, P., Isci, C., Kalantar, M., Segmuller, W., & Snible, E. C. (2016) · IBM Journal of Research and Development · doi:10.1147/jrd.2016.2517498 · Cited by 16
Markov Chain Modeling for Anomaly Detection in High Performance Computing System Logs
Haque, A., DeLucia, A., & Baseman, E. (2017) · International Workshop on HPC User Support Tools (HUST) · doi:10.1145/3152493.3152559 · Cited by 16
Latent Variable Based Anomaly Detection in Network System Logs
Otomo, K., Kobayashi, S., Fukuda, K., & Esaki, H. (2019) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2018ofp0007 · Cited by 17
“Some latent analysis methods define the anomaly level based on reconstruction errors [18], [21].”
MethodIdentifying faults in large-scale distributed systems by filtering noisy error logs
Rao, X., Wang, H., Shi, D., Chen, Z., Cai, H., Zhou, Q., & Sun, T. (2011) · International Conference on Dependable Systems and Networks workshops · doi:10.1109/dsnw.2011.5958800 · Cited by 16
“A common way to tackle this problem is to use text extraction techniques [10].”
Xie, X., Jian, S., Huang, C., Yu, F., & Deng, Y. (2023) · Proceedings/Proceedings - International Symposium on Software Reliability Engineering · doi:10.1109/issre59848.2023.00015 · Cited by 18
“Many traditional methods used in classification problems are utilized to solve log-based anomaly detection, such as Support Vector Machine (SVM) [24], Invariant Mining (IM) [26], Principal Component Analysis (PCA) [46, 16].”
BackgroundMethodInfluentialQiu, K., Zhang, Y., Feng, Y., & Chen, F. (2025) · Journal of Network and Systems Management · doi:10.1007/s10922-025-09912-5 · Cited by 15
“HDFS [45, 46] This log set was collected by aggregating logs from an HDFS system in the labs of the Chinese University of Hong Kong, which contains a name node and 32 data nodes.”
SemiRALD: A semi-supervised hybrid language model for robust Anomalous Log Detection
Sun, Y., Keung, J., Yang, Z., Liu, S., & Yu, H. K. (2025) · Information and Software Technology · doi:10.1016/j.infsof.2025.107743 · Cited by 18
BigDebug: interactive debugger for big data analytics in Apache Spark
Gulzar, M. A., Interlandi, M., Condie, T., & Kim, M. (2016) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/2950290.2983930 · Cited by 15
“Xu et al. parse console logs and combine source code analysis to detect abnormal behavior [12].”
BackgroundOnline Reconstruction of Structural Information from Datacenter Logs
Chothia, Z., Liagouris, J., Dimitrova, D., & Roscoe, T. (2017) · European Conference on Computer Systems · doi:10.1145/3064176.3064195 · Cited by 19
“All these tasks are based on the output of sessionization, and they could be easily composed to form more complex tasks [28, 42, 44, 52].”
BackgroundMethodInfluentialYuan, Y., Anu, H., Shi, W., Liang, B., & Qin, B. (2019) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac.2019.00019 · Cited by 15
“Log parsing has been widely studied recently [7], [12], [16], [32].”
BackgroundMethodAnomaly Detection in Distributed Systems via Variational Autoencoders
Qian, Y., Ying, S., & Wang, B. (2020) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc42975.2020.9283078 · Cited by 19
“In addition to supervised methods, many unsupervised anomaly detection methods have been proposed [9]–[11].”
BackgroundMethodInfluentialMulti-source Anomaly Detection in Distributed IT Systems
Bogatinovski, J., & Nedelkoski, S. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-76352-7_22 · Cited by 18
FlexParser—The adaptive log file parser for continuous results in a changing world
Rücker, N., & Maier, A. (2022) · Journal of Software Evolution and Process · doi:10.1002/smr.2426 · Cited by 15
“sult: learn to map this entire string to 456 (1) ”the interesting value =123,some other text” ”another value =456,some more text” →result: learn to extract the number between ’value’ and ’some’ (2) 2.3.3 Fully Convolutional Network Anotherstate-of-the-arttimeseriesclassificationmethodisrepresentedbyfullyconvolutionalne…”
BackgroundInfluentialWang, Z., Zhang, H., Chen, T.-H. P., & Wang, S. (2021) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3468264.3468613 · Cited by 19
“Then, developers use the generated logs to assist in monitoring [13, 40], testing [6, 7], and debugging [5, 42, 43].”
BackgroundMethodLog Anomaly Detection method based on BERT model optimization
Shao, Y., Zhang, W., Liu, P., Ren, H., Tang, R., Yin, Q., & Li, Q. L. (2022) · International Conference on Cloud Computing and Big Data Analytics (ICCCBDA) · doi:10.1109/icccbda55098.2022.9778900 · Cited by 20
SKDLog: Self-Knowledge Distillation-based CNN for Abnormal Log Detection
Han, N., Lu, S., Wang, D., Wang, M., Tan, X., & Wei, X. (2022) · IEEE Smart World Congress (SWC) · doi:10.1109/smartworld-uic-atc-scalcom-digitaltwin-pricomp-metaverse56740.2022.00122 · Cited by 18
“Their detailed information is introduced as follows: HDFS. HDFS [36] log dataset is produced from more than 200 Amazon EC2 nodes for Hadoop distributed file system in runtime, which collects 11,175, and 629 logs, plus 29 unique log keys.”
MethodSCMiner: Localizing System-Level Concurrency Faults from Large System Call Traces
Zaman, T. S., Han, X., & Yu, T. (2019) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1109/ase.2019.00055 · Cited by 15
“There has been some research on detecting anomalies [33], [54], [55] from logs.”
BackgroundCloudRaid: Detecting Distributed Concurrency Bugs via Log Mining and Enhancement
Lu, J., Li, F., Liu, C., Li, L., Feng, X., & Xue, J. (2020) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2020.2999364 · Cited by 15
“We adopt the approach in [17] to match each runtime log instance with a static message logging expression efficiently.”
BackgroundMethodInfluentialOrca: differential bug localization in large-scale services
(2018) · Operating Systems Design and Implementation · doi:10.5555/3291168.3291205 · Cited by 15
An Efficient Log Parsing Algorithm Based on Heuristic Rules
Zhang, L., Xie, X., Xie, K., Wang, Z., Lu, Y., & Zhang, Y. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-29611-7_10 · Cited by 15
Symptom-based problem determination using log data abstraction
Huang, L. P., Ke, X., Wong, K., & Mankovskii, S. (2010) · IEEE International Conference on Collaborative Advances in Software and Computing (CASCON) · doi:10.1145/1923947.1923979 · Cited by 18
“And approaches that extract the line patterns through source code analysis like [36] are also not suitable for the same reason.”
BackgroundSelf-citationThe Insider Threat Detection Method of University Website Clusters Based on Machine Learning
Li, Y., & Su, Y. (2023) · 2023 6th International Conference on Artificial Intelligence and Big Data (ICAIBD) · doi:10.1109/icaibd57115.2023.10206282 · Cited by 14
“using PCA [19] to detect large-scale system issues by mining console logs, Lou et al.”
MethodPerformance Bug Analysis and Detection for Distributed Storage and Computing Systems
Li, J., Zhang, Y., Lu, S., Gunawi, H. S., Gu, X., Huang, F., & Li, D. (2023) · ACM Transactions on Storage · doi:10.1145/3580281 · Cited by 14
“Many tools have been built to diagnose system performance anomalies [7, 9, 16, 46ś48, 53], including many built for diagnosing and reasoning about performance problems in distributed systems [5, 11, 38, 49, 57, 64].”
Metadata
- Authors
- Xu, W., Huang, L., Fox, A., Patterson, D., & Jordan, M. I.
- Affiliations
- University of California, Berkeley (US)
- Intel (United States) (US)
- Venue
- Proceedings of the ACM SIGOPS 22nd symposium on Operating systems principles
- Publisher
- ACM
- Published
- 2009-10-11
- Pages
- 117-132
- OpenAlex
- W2039157918
- Cited by
- 1,332
- References
- 44
- Citations by year
- 2026: 100 · 2025: 126 · 2024: 134 · 2023: 138 · 2022: 97 · 2021: 126 · 2020: 96 · 2019: 89 · 2018: 72 · 2017: 49 · 2016: 62 · 2015: 48 · 2014: 41 · 2013: 37 · 2012: 42
- FWCI
- 31.53
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- ACM SIGOPS Symposium on Operating Systems Principles (SOSP) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Software System Performance and Reliability
- Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- Physical Sciences › Computer Science › Software › Software Testing and Debugging Techniques
- Keywords
- fault detection, log parsing, composite features, source code analysis, information retrieval, decision tree, Hadoop
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.