MétaCan
Menu
Back to cohort

Efficient Detection of Communication-related Performance Anti-patterns in Microservices

2025· article· en· W4413377439 on OpenAlexafffund
Masoumeh Nourollahi, Naser Ezzati‐Jivan, Adel Belkheiri, Michel Dagenais

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicCloud Computing and Resource Management
Canadian institutionsBrock UniversityPolytechnique Montréal
FundersNatural Sciences and Engineering Research Council of Canada
KeywordsMicroservicesComputer scienceOperating system

Abstract

fetched live from OpenAlex

Modern microservice-based applications are inherently complex due to their distributed nature and intricate service interactions, making performance diagnosis challenging. A significant source of degradation in such systems is Software Performance Anti-patterns, which can arise from inefficient coding practices, poor architectural design, or suboptimal deployment strategies. Existing Anti-pattern detection methods often rely on costly, intrusive data collection, limiting their practicality in production environments. We propose a low-cost, non-intrusive approach for detecting communication-related Anti-patterns by combining selectively captured networking-related system call traces with distributed observability traces. This hybrid tracing enables detailed correlation between low-level communication metrics and high-level service interactions, supporting accurate detection of Anti-patterns such as Blob and Empty-semi-trucks. To minimize overhead, only essential system call events are collected, while distributed tracing data is used for root cause analysis at the service and operation level. We design a machine learning pipeline, supporting super- vised, semi-supervised, and unsupervised detection, achiev- ing up to 91% accuracy with just 2.74% data collection over- head. Offline model training further supports seamless in- tegration into Continuous Integration/Continuous Deploy- ment (CI/CD) workflows for early performance regression detection. Our approach is validated on the DeathStarBench microservice benchmark across 14 scenarios under both clean and noisy conditions. Results show high detection accuracy, effective root cause identification (over 80% match to manual analysis), and minimal runtime impact. This work offers a practical, scalable, and accurate solution for SPA detection in complex microservice environments. Keywords: software performance anti-patterns, tracing, system calls, kernel tracing, distributed tracing, semi-supervised learning

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.005
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.004
Threshold uncertainty score0.008

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.005
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.006
GPT teacher head0.218
Teacher spread0.213 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes2
Has abstractyes

Explore more

Same topicCloud Computing and Resource ManagementFrench-language works237,207