Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM65465
Volume:
10 |
6 |
Jun 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
Department of Computer Engineering
Sinhgad College of Engineering
The proliferation of polyglot microservice architectures has created an observability crisis in which traditional threshold-based monitoring systems fail to provide automated causal reasoning for distributed failures. This paper introduces PulseX, a novel Cognitive Observability framework that combines zero-code auto-instrumentation via OpenTelemetry (OTel), stateful stream processing with Bytewax, a five-model ensemble machine-learning anomaly scoring engine, and structured Generative AI Root Cause Analysis (RCA) powered by Gemini 2.5 Flash Lite. Cross-service traces are reconstructed within a 10- second tumbling window and evaluated against four complementary anomaly detectors: N+1 query regression, bimodal latency, dependency chain break, and PII redaction density anomaly—the latter constituting a novel security primitive not found in prior literature. The ensemble (Isolation Forest, One Class SVM, Autoencoder, HS-Trees, LOF) is validated on both a high-fidelity synthetic telemetry corpus and the LO2 real-world microservice anomaly dataset. Evaluation demonstrates an 88.7% F1-score on synthetic traces, 97.5% recall on the LO2 dataset, a 95.2% average RCA precision across 12 live Geminievaluated fault scenarios, and a 20–180× improvement in timeto-awareness over human-in-the-loop monitoring. Index Terms—OpenTelemetry, Bytewax, River ML, Root Cause Analysis, Generative AI, Microservices Observability, Anomaly Detection, Cognitive Observability, AIOps.
KEYWORDS
PulseX demonstrates that a carefully composed stack of zero-code instrumentation, Rust-powered stateful streaming, online ensemble ML, and structured LLM reasoning eliminates the ”detection-to-diagnosis” gap in modern microservice operations. The system achieves 93.7% F1-score anomaly detection and 96.4% RCA precision with sub-12-second end-to-end latency on a commodity 8GB development machine—without Kubernetes, Elasticsearch, or a JVM. The PII Redaction Density detector constitutes an original contribution: a novel security primitive derived from telemetry infrastructure itself. We position Cognitive Observability as a new research agenda at the intersection of AIOps, streaming data management, and explainable AI.
1) V. Chandola, A. Banerjee, and V. Kumar, “Anomaly detection: A survey,” ACM Comput. Surveys, vol. 41, no. 3, pp. 1–58, 2009. 2) Uber Engineering, “Jaeger: Open source, end-to-end distributed tracing,” CNCF, 2017. [Online]. Available: https://www.jaegertracing.io 3) Twitter Engineering, “Zipkin: A distributed tracing system,” Opensourced 2012. [Online]. Available: https://zipkin.io 4) Grafana Labs, “Grafana Tempo: A distributed tracing backend,” 2020. [Online]. Available: https://grafana.com/oss/tempo/ 5) F. McSherry, M. Isard, and D. Murray, “Naiad: A timely dataflow system,” in Proc. 24th ACM SOSP, 2013, pp. 439–455. 6) M. Panahandeh, A. Hamou-Lhadj, M. Hamdaqa, and J. Miller, “ServiceAnomaly: An anomaly detection approach in microservices using distributed traces and profiling metrics,” J. Syst. Softw., vol. 209, 2024. 7) T. Ahmed et al., “Recommending root-cause and mitigation steps for cloud incidents using large language models,” in Proc. ICSE, 2023, pp. 1737–1749. 8) Z. Wang et al., “RCAgent: Cloud root cause analysis by autonomous agents with tool-augmented LLMs,” in Proc. CIKM, 2024. 9) A. Spadaccini and K. Guliani, “AIOps for reliability: Evaluating LLMs for automated RCA in chaos engineering,” in Proc. ICCS, 2025. 10) M. Halford, “River: Online machine learning in Python,” riverml.xyz, 2021. [Online]. Available: https://riverml.xyz
Publication Date
6/25/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
4
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...