A Comprehensive Benchmark and Empirical Study of Trace Anomaly Detection
The DOI matches its Crossref record; 22 of the 23 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
The growing complexity of modern Internet applications and the widespread use of microservice architectures have amplified the need for efficient trace anomaly detection to maintain system stability. Despite the fact that many trace anomaly detection algorithms have been proposed to identify abnormal behaviors, a comprehensive evaluation of these methods is lacking, which makes it difficult for developers to choose the most suitable algorithm for real-world applications. To address this gap, we presentTADBench, a comprehensive and extensible benchmark for trace anomaly detection.TADBenchconsolidates diverse publicly available trace datasets and algorithms into a unified repository, standardizes data formats, and incorporates manual anomaly labels. To ensure reproducibility and fair comparisons, we propose a modular evaluation framework supporting end-to-end model assessment. Additionally, we provide practical guidance for algorithm selection based on specific data attributes by evaluating their performance across datasets with different characteristics, thereby effectively bridging the gap between academic research and industrial deployment. To the best of our knowledge, this is the first comprehensive empirical study of trace anomaly detection algorithms. Our findings aim to facilitate the adoption of these methods in production environments, offering actionable insights for developers and researchers.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“As models may reproduce memorized content, operational agents should, by default, avoid persistent memory and treat each incident as a distinct retention boundary [154], [155], [156].”
From Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety (2027) · cites this paper as [154]
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1×
“3) LSTM-Based Algorithms: Long Short-Term Memory (LSTM) networks [33] are used in this category to model sequential dependencies and detect both latency and structural anomalies in traces.”
Classification And Regression Trees
Breiman, L., Friedman, J. H., Olshen, R. A., & Stone, C. J. (2017) · doi:10.1201/9781315139470 · Cited by 11,417
DeepWalk: online learning of social representations
Perozzi, B., Al-Rfou, R., & Skiena, S. (2014) · Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/2623330.2623732 · Cited by 10,123 · cited 1×
“Initially, it utilizes Deep-Walk [31] to generate vector representations for each operation.”
Microservices: Yesterday, Today, and Tomorrow
Dragoni, N., Giallorenzo, S., Lafuente, A. L., Mazzara, M., Montesi, F., Mustafin, R., & Safina, L. (2017) · Springer eBooks · doi:10.1007/978-3-319-67425-4_12 · Cited by 2,058
An End-to-End Deep Learning Architecture for Graph Classification
Zhang, M., Cui, Z., Neumann, M., & Chen, Y. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11782 · Cited by 1,461
Japanese and Korean voice search
Schuster, M., & Nakajima, K. (2012) · 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2012.6289079 · Cited by 862 · cited 1×
“Notably, when generating graph representations for traces,PUTraceADusesWordPiece[27]andapre-trainedBERT model[28]togenerate768-dimensionalembeddingsforservice andoperationnameswhilecreatinggraphrepresentationsfor traces.”
GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders
Simonovsky, M., & Komodakis, N. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01418-6_41 · Cited by 563
Microservices: The Journey So Far and Challenges Ahead
Jamshidi, P., Pahl, C., Mendonca, N. C., Lewis, J., & Tilkov, S. (2018) · IEEE Software · doi:10.1109/ms.2018.2141039 · Cited by 554
Zhou, X., Peng, X., Xie, T., Sun, J., Ji, C., Li, W., & Ding, D. (2021) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2018.2887384 · Cited by 410 · cited 1×
“In this study, we conduct an in-depth analysis of multiple datasets, including TrainTicket [35], [36], GAIA (Generic AIOps Atlas) [37], AIOps2020 [38], AIOps2022 [39] and AIOps2023.”
Villamizar, M., Garces, O., Castro, H., Verano, M., Salamanca, L., Casallas, R., & Gil, S. (2015) · 2015 10th Computing Colombian Conference (10CCC) · doi:10.1109/columbiancc.2015.7333476 · Cited by 308 · cited 2×
“This architecture enables developers and engineers to independently develop, deploy and update services, greatly boosting development efficiency and system responsiveness [1].”
Unsupervised Detection of Microservice Trace Anomalies through Service-Level Deep Bayesian Networks
Liu, P., Xu, H., Ouyang, Q., Jiao, R., Chen, Z., Zhang, S., Yang, J., Mo, L., Zeng, J., Xue, W., et al. (2020) · 2020 IEEE 31st International Symposium on Software Reliability Engineering (ISSRE) · doi:10.1109/issre5003.2020.00014 · Cited by 252 · cited 1×
“It enables the real-time identification of issues within the service interaction network, facilitating the prompt localization and resolution of faults and thereby preventing service interruptions or performance degradation [4].”
Self-citationDeepTraLog: trace-log combined microservice anomaly detection through graph-based deep learning
Zhang, C., Peng, X., Sha, C., Zhang, K., Fu, Z., Wu, X., Lin, Q., & Zhang, D. (2022) · Proceedings of the 44th International Conference on Software Engineering · doi:10.1145/3510003.3510180 · Cited by 207 · cited 1×
“Furthermore, historical trace analysis can reveal long-term performance trends and abnormal patterns, offering crucial insights for system optimization and future architectural improvements [7].”
Benchmarking microservice systems for software engineering research
Zhou, X., Peng, X., Xie, T., Sun, J., Xu, C., Ji, C., & Zhao, W. (2018) · Proceedings of the 40th International Conference on Software Engineering: Companion Proceeedings · doi:10.1145/3183440.3194991 · Cited by 168
(2017) · doi:10.1145/3132747.3132749 · Cited by 144 · cited 1×
“Developers can gain a comprehensive understanding of service interaction patterns by monitoring the dependencies and sequences of service calls, which allows them to optimize service performance and improve system stability [5], [6].”
Anomaly Detection from System Tracing Data Using Multimodal Deep Learning
Nedelkoski, S., Cardoso, J., & Kao, O. (2019) · 2019 IEEE 12th International Conference on Cloud Computing (CLOUD) · doi:10.1109/cloud.2019.00038 · Cited by 139
Graph-based trace analysis for microservice architecture understanding and problem diagnosis
Guo, X., Peng, X., Wang, H., Li, W., Jiang, H., Ding, D., Xie, T., & Su, L. (2020) · Proceedings of the 28th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering · doi:10.1145/3368089.3417066 · Cited by 118
Kwon, Y., Yoo, J., Choi, Y.-S., Son, W., Lee, D., & Kang, S. (2019) · Journal of Cheminformatics · doi:10.1186/s13321-019-0396-x · Cited by 89
Improving Tree-LSTM with Tree Attention
Ahmed, M., Samee, M. R., & Mercer, R. E. (2019) · IEEE International Conference on Semantic Computing (ICSC) · doi:10.1109/icosc.2019.8665673 · Cited by 67
Unsupervised Anomaly Detection on Microservice Traces through Graph VAE
Xie, Z., Xu, H., Chen, W., Li, W., Jiang, H., Su, L., Wang, H., & Pei, D. (2023) · Proceedings of the ACM Web Conference 2023 · doi:10.1145/3543507.3583215 · Cited by 54 · cited 5×
“(cid:2) Towards fine-grained span-level anomaly detection: Many anomaly detection algorithms [14], [16], [20], [25], [34] operate primarily at the trace level, labeling an entire trace as normal or abnormal.”
InfluentialSelf-citationTraceCRL: contrastive representation learning for microservice trace analysis
Zhang, C., Peng, X., Zhou, T., Sha, C., Yan, Z., Chen, Y., & Yang, H. (2022) · Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering · doi:10.1145/3540250.3549146 · Cited by 36 · cited 2×
“TraceCRL: Following the GraphCL (Graph Contrastive Learning with Augmentations) framework [29], TraceCRL [30] leverages contrastive learning and graph neural networks to learn effective trace representations.”
PUTraceAD: Trace Anomaly Detection with Partial Labels based on GNN and PU Learning
Zhang, K., Zhang, C., Peng, X., & Sha, C. (2022) · 2022 IEEE 33rd International Symposium on Software Reliability Engineering (ISSRE) · doi:10.1109/issre55969.2022.00032 · Cited by 33 · cited 3×
“TraceVAE [20] demonstrates leading performance on the GAIA and AIOps2022 datasets with F1-scores of 90.9% and 78.9% respectively, while PUTraceAD [25] achieves the highest F1-score of 74.7% on the AIOps2023 dataset.”
InfluentialFrom Point-wise to Group-wise: A Fast and Accurate Microservice Trace Anomaly Detection Approach
Xie, Z., Pei, C., Li, W., Jiang, H., Su, L., Li, J., Xie, G., & Pei, D. (2023) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3611643.3613861 · Cited by 28
Self-citationEfficient and Robust Trace Anomaly Detection for Large-Scale Microservice Systems
Zhang, S., Pan, Z., Liu, H., Jin, P., Sun, Y., Ouyang, Q., Wang, J., Jia, X., Zhang, Y., Yang, H., et al. (2023) · 2023 IEEE 34th International Symposium on Software Reliability Engineering (ISSRE) · doi:10.1109/issre59848.2023.00012 · Cited by 14
Self-citation
Research works printed but not linked: 11
- Kingma, 2014, Auto-encoding variational Bayes, 1
- 10.1109/issre5003.2020.00014 · DOI · Paper page
- Rezende, 2015, Variational inference with normalizing flows, 1530
- Zhang, 2022, CRISP: Critical path analysis of large-scale microservice architectures, 655
- Mitton, 2020, A graph VAE and graph transformer approach to generating molecular graphs, 1
- Veličković, 2018, Graph attention networks, 1
- Hamilton, 2017, Inductive representation learning on large graphs, 1025
- Kiryo, 2017, Positive-unlabeled learning with non-negative risk estimator, 1674
- Wolf, 2020, Transformers: State-of-the-art natural language processing, 38
- You, 2020, Graph contrastive learning with augmentations, 33, 5812
- Schölkopf, 1999, Support vector method for novelty detection, 12, 582
Cited by
Large language models for agentic NetOps and AIOps: Architectures, evaluation, and safety
Bilal, M., Crowcroft, J., Wang, R., Xu, X., & Dustdar, S. (2027) · Computer Science Review · doi:10.1016/j.cosrev.2026.101075 · Cited by 10
“This matters when agents must decide whether an anomaly requires escalation or mitigation [136].”
Zhang, X., Wang, Z., Wang, P., Cao, Z., & Liu, M. (2026) · KSII Transactions on Internet and Information Systems · doi:10.3837/tiis.2026.06.014
CodedFRER: Enhancing Reliability for FRER Protocol via Erasure Codes in Error-Prone Networks
Dun, S., Jia, H., Gan, C., Hu, Y., Long, T., & Feng, D. (2026) · International Workshop on Quality of Service · doi:10.1109/iwqos70441.2026.11661014
“We set σ in Gaussian kernel to R tar / 3 based on the 3 σ rule [26], ensuring W gain decays to negligible values ( ≈ 0 .”
Intelligent Learning Model for Anomaly Detection in Microservice Architecture
Tun, T., Intana, A., & Tantayakul, K. (2026) · Computers · doi:10.3390/computers15110770
Metadata
- Authors
- Sun, Y., Shao, M., Nie, X., Yang, K., Li, X., Hao, B., Zhang, S., Pei, C., He, D., Li, Y., & Pei, D.
- Affiliations
- Nankai University (CN)
- Chinese Academy of Sciences (CN)
- Tsinghua University (CN)
- ORCID
- Venue
- IEEE Transactions on Services Computing
- Published
- 2025-11
- Volume
- 18
- Issue
- 6
- Pages
- 3364-3377
- ISSN
- 1939-1374 (ISSN-L)
- 2372-0204
- OpenAlex
- W4415256963
- Cited by
- 1
- References
- 23
- Citations by year
- 2026: 1
- FWCI
- 0.42
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Services Computing (publishedVersion)
- Funders
- National Natural Science Foundation of China (62272249, 62302244)
- Fundamental Research Funds for the Central Universities (XXX-63253249)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Software System Performance and Reliability
- Physical Sciences › Computer Science › Artificial Intelligence › Anomaly Detection Techniques and Applications
- Physical Sciences › Computer Science › Computer Networks and Communications › Network Security and Intrusion Detection
- Keywords
- benchmark, microservices, distributed tracing, anomaly detection algorithms, algorithm evaluation, reproducibility, algorithm selection
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.