{
  "schema_version": "0.6.0",
  "paper_id": "efficient-unsupervised-latency-culprit-ranking",
  "page_url": "https://naser.github.io/research-publications/papers/efficient-unsupervised-latency-culprit-ranking/",
  "title": "Efficient Unsupervised Latency Culprit Ranking in Distributed Traces with GNN and Critical Path Analysis",
  "title_variants": [],
  "authors": [
    "Mahsa Panahandeh",
    "Naser Ezzati-Jivan",
    "Abdelwahab Hamou-Lhadj",
    "James Miller"
  ],
  "author_details": [
    {
      "name": "Mahsa Panahandeh",
      "orcid": "https://orcid.org/0000-0002-6369-8982",
      "profile_url": "https://dblp.org/pid/201/1655.html"
    },
    {
      "name": "Naser Ezzati-Jivan",
      "orcid": "https://orcid.org/0000-0003-1435-6297",
      "profile_url": "https://naser.github.io/"
    },
    {
      "name": "Abdelwahab Hamou-Lhadj",
      "orcid": "https://orcid.org/0000-0002-3319-5006",
      "profile_url": "https://dblp.org/pid/70/2136.html"
    },
    {
      "name": "James Miller",
      "orcid": "https://orcid.org/0000-0001-5095-3000",
      "profile_url": "https://dblp.org/pid/44/7043-1.html"
    }
  ],
  "publication": {
    "year": 2024,
    "venue": "ACM/SPEC International Conference on Performance Engineering (ICPE) Companion",
    "type": "conference paper",
    "publication_date": "2024-05-07",
    "online_date": "2024-05-07",
    "print_date": "2024-05-07",
    "volume": null,
    "issue": null,
    "pages": "62-66",
    "article_number": null,
    "publisher": "ACM",
    "issn": [],
    "isbn": [],
    "crossref_type": "proceedings-article"
  },
  "publication_type": "conference paper",
  "status": "published_with_public_full_text",
  "canonical_source_url": "https://doi.org/10.1145/3629527.3651841",
  "source_record_id": "efficient-unsupervised-latency-culprit-ranking-in-distributed-traces-with-gnn-and-critical-path--ec5a0cb026",
  "identifiers": {
    "doi": "10.1145/3629527.3651841"
  },
  "abstract": null,
  "abstract_source": "Abstract not reproduced; the four-part summary is paraphrased from the verified full text.",
  "abstract_available": false,
  "scholar_eligibility": {
    "eligible": false,
    "basis": "not-eligible",
    "note": "The page is a discovery record; it does not claim Google Scholar article-host eligibility."
  },
  "description": "The paper combines an unsupervised GraphSAGE model with critical-path-specific latency profiles to detect anomalous requests and rank likely microservice culprits without labelled training data.",
  "evidence_level": "full-text-reviewed",
  "evidence": {
    "source_basis": "full-text-reviewed",
    "coverage": "material paper sections",
    "summary_origin": "AI-assisted catalog editorial summary",
    "review_status": "catalog-reviewed; paper-author approval pending",
    "verified_on": "2026-08-09",
    "sources": [
      {
        "note": "Private full PDF read: pdf-evidence/notes/efficient-unsupervised-latency-culprit-ranking.md"
      },
      {
        "note": "Efficient culprit-ranking PDF pp. 2-4: distributed-trace input boundary, GraphSAGE model, MSE threshold, critical-path profiles, FIRM dataset, and split"
      },
      {
        "note": "Efficient culprit-ranking PDF pp. 4-5: benchmark accuracy, Top-k results, training/test speed comparison, limitations, and artifact link"
      },
      {
        "note": "Local PDF hash verified in pdf-evidence/extraction-manifest.json"
      }
    ]
  },
  "summary": {
    "core_contribution": "The paper combines an unsupervised GraphSAGE model with critical-path-specific latency profiles to detect anomalous requests and rank likely microservice culprits without labelled training data.",
    "problem": "Latency-culprit methods for microservices often require labeled anomalies or make assumptions about propagation paths that do not hold across requests. The paper seeks request-level culprit ranking without a labeled training set.",
    "method": "Historical service-span latency vectors are normalized and grouped by computed critical path. Each critical-path cluster stores service latency distributions. A two-layer GraphSAGE encoder/decoder reconstructs node features on a directed acyclic service-invocation graph; MSE above 0.1 marks an anomalous request. For each candidate service, a sampled latency from its historical distribution replaces the observed value; candidates whose replacement makes the request normal are retained and ranked by distributional deviation. A static graph plus periodic updates and neighborhood sampling is used for efficiency.",
    "findings": "On FIRM preprocessed traces for social-network, hotel-reservation, media-service, and ticket-booking benchmarks, ACC is 87%, 95.4%, 86.4%, and 96%; Top-1 is 83%, 94.7%, 85%, and 94.3%; Top-3 is 86.2%, 95.4%, 86.4%, and 96%; Top-5 is 87%, 95.4%, 86.4%, and 96%, respectively. The paper reports roughly 3-8% accuracy improvement and training-time reduction to more than one-fifth of a comparable method; the critical-path version reduces test culprit-identification time from 1.5-13 s to 0.8-8.3 s, reported as a 58.33% average improvement.",
    "limitations": "Static service graphs require updates; a computed critical path can overshadow the actual culprit in sparse systems or when the culprit has a narrow latency distribution. The evaluation uses aggregated, preprocessed FIRM data rather than arbitrary raw tracing backends, and the paper does not establish production transfer.",
    "future_work": "Broaden experiments, improve detection, handle evolving dependencies, and test larger or more varied service graphs and deployment conditions."
  },
  "tags": [
    "microservices",
    "graph-neural-networks",
    "latency-analysis",
    "root-cause-analysis",
    "trace-analysis"
  ],
  "keywords": [
    "latency culprit ranking",
    "distributed traces",
    "GraphSAGE",
    "graph neural networks",
    "critical path",
    "FIRM dataset",
    "service invocation graph",
    "unsupervised anomaly detection",
    "Top-k ranking"
  ],
  "versions": [
    {
      "id": "published-version",
      "label": "Published version",
      "relation": "version-of-record",
      "title": "Efficient Unsupervised Latency Culprit Ranking in Distributed Traces with GNN and Critical Path Analysis",
      "url": "https://doi.org/10.1145/3629527.3651841",
      "pdf_url": null,
      "status": "published",
      "canonical_for_citation": true
    },
    {
      "id": "public-acm-pdf",
      "label": "Public ACM PDF",
      "relation": "source-record",
      "title": "Efficient Unsupervised Latency Culprit Ranking in Distributed Traces with GNN and Critical Path Analysis",
      "url": "https://dl.acm.org/doi/pdf/10.1145/3629527.3651841",
      "pdf_url": "https://dl.acm.org/doi/pdf/10.1145/3629527.3651841",
      "status": "public_full_text",
      "canonical_for_citation": false
    }
  ],
  "access": {
    "status": "published_with_public_full_text",
    "note": "The DOI is the canonical citation target. The external source PDF is linked for readers; no PDF is redistributed in this repository.",
    "license": null
  },
  "resources": {
    "code": "https://anonymous.4open.science/r/ICPE2024-4281/v2.py",
    "data": "https://doi.org/10.13012/B2IDB-6738796_V1",
    "slides": null,
    "demo": null
  },
  "citation_guidance": {
    "when_to_cite": "Cite this paper when your work uses or compares an unsupervised GraphSAGE encoder/decoder that detects request-level latency anomalies from service-span vectors.",
    "points": [
      "an unsupervised GraphSAGE encoder/decoder that detects request-level latency anomalies from service-span vectors.",
      "critical-path clustering plus service latency distributions as a label-free culprit-ranking strategy.",
      "the FIRM benchmark ACC/Top-k table or the v1/v2 timing comparison when evaluating efficient distributed-trace culprit localization.",
      "with the static-graph and preprocessed-data caveat when discussing production generalization."
    ],
    "canonical_version_id": "published-version"
  },
  "provenance": {
    "metadata_verified_on": "2026-08-09",
    "metadata_source": [
      "Private full PDF read: pdf-evidence/notes/efficient-unsupervised-latency-culprit-ranking.md",
      "Efficient culprit-ranking PDF pp. 2-4: distributed-trace input boundary, GraphSAGE model, MSE threshold, critical-path profiles, FIRM dataset, and split",
      "Efficient culprit-ranking PDF pp. 4-5: benchmark accuracy, Top-k results, training/test speed comparison, limitations, and artifact link",
      "Local PDF hash verified in pdf-evidence/extraction-manifest.json"
    ],
    "summary_written_by": "AI-assisted",
    "summary_verified_by": "full-text-grounded catalog review; author approval pending",
    "linked_preprint_record": null,
    "author_order_note": null
  },
  "batch": {
    "phase": 2,
    "batch_label": "expanded forty-paper release",
    "status": "included_in_expanded_catalog",
    "selected_at": "2026-08-09"
  }
}