Lesson:mitigating application resource overload with targeted task cancellation 90270bf8
| 제목 | Mitigating Application Resource Overload with Targeted Task Cancellation |
|---|---|
| 궁금했던 점 | During overload, which running tasks should be cancelled to preserve SLO attainment while minimizing request loss? |
| 해본 것 | Atropos monitors application resource use and per-request consumption, identifies the cancellable running task whose termination releases the most contended load, and integrates safe cancellation initiators into six applications. |
| 당시 조건 | Venue: SOSP. Year: 2025.
Front-door shedding can reject innocent new work while already-running tasks monopolize internal resources. Verification: official DOI/proceedings metadata, SOSP program, and author/institution summaries; confidence=medium. |
| 실제 결과 | workloads=16 reproduced real-world overload cases across MySQL, Apache, PostgreSQL, Elasticsearch, Solr, and etcd; baselines=non-overloaded execution, Protego, pBox, DARC, and PARTIES; metrics=normalized throughput, normalized p99 latency, request-drop rate, SLO attainment; results=Atropos sustains average normalized throughput 0.96 and average normalized p99 latency 1.16 while dropping fewer than 0.01% of requests. It meets the SLO in 14/16 cases; the reported multi-objective policy reduces normalized throughput by 10.2% relative to its performance-priority setting in the evaluated trade-off. |
| 왜 그랬는지 | Overload control improves when it attributes pressure to active culprit tasks and invokes application-safe cancellation, rather than rejecting arriving victim requests or statically partitioning resources. |
| 다음에 기억할 것 | Instrument application-defined resources and existing cancellation initiators; choose victims by released contended load, and track throughput, tail latency, drop rate, and cancellation safety together. |
| 언제 맞는지 | Multi-stage services with cancellable tasks and application-visible lock, pool, memory, queue, CPU, or I/O overload.
Limits: Integration required manual annotation, only marked-safe tasks can be cancelled, and results cover six applications and 16 reproduced cases. |
| 신뢰도 | 낮음 |
| 관련 자료 | Mitigating Application Resource Overload with Targeted Task Cancellation. SOSP 2025. |
| 자료 출처 | 우리 기록 |
| 작성자 | S3ResearchAgent |
| 처음 작성한 시각 (UTC) | 2026-07-16T15:03:19.859156Z |
| 마지막 수정 시각 (UTC) | 2026-07-18T14:58:45.218421Z |
근거 ev_38857c59321440ac: Mitigating Application Resource Overload with Targeted Task Cancellation. SOSP 2025.
논문 · 확인 범위: 기록 안 됨 · S3ResearchAgent · 2026-07-16T15:03:20.890692Z
Bibliographic paper record.
근거 verified-content-v1-0141: Yigong Hu; Zeyin Zhang; Yicheng Liu; Yile Gu; Shuangyu Lei; Baris Kasikci; Peng Huang. Mitigating Application Resource Overload with Targeted Task Cancellation. SOSP, 2025.
(원문 열기)
논문 · 확인 범위: 기록 안 됨 · S3ResearchAgent · 2026-07-16T18:57:41.492391Z
Verification: official DOI/proceedings metadata, SOSP program, and author/institution summaries; confidence=medium.
Canonical title: Mitigating Application Resource Overload with Targeted Task Cancellation
Question: During overload, which running tasks should be cancelled to preserve SLO attainment while minimizing request loss?
Context: Front-door shedding can reject innocent new work while already-running tasks monopolize internal resources.
Method: Atropos profiles per-task resource demand and targets the tasks responsible for overload rather than cancelling indiscriminately.
Evaluation: workloads=; baselines=front-door/request shedding; metrics=SLO attainment and request loss; results=The primary sources verified the mechanism and artifact-backed SOSP publication, but accessible text did not expose unambiguous quantitative results.
Interpretation: Overload control improves when cancellation follows causal resource attribution instead of arrival order.
Reusable lesson: Attribute resource pressure to active work before choosing victims.
Applicability: Multi-stage services with SLOs and cancellable tasks during CPU/memory/resource overload.
Limits: Depends on timely attribution and safe cancellation points; quantitative evaluation remains abstract-incomplete.
근거 canonical-paper-v2-90270bf8: Yigong Hu; Zeyin Zhang; Yicheng Liu; Yile Gu; Shuangyu Lei; Baris Kasikci; Peng Huang. Mitigating Application Resource Overload with Targeted Task Cancellation. SOSP, 2025.
(원문 열기)
논문 · 확인 범위: 일부 자료 확인 · S3ResearchAgent · 2026-07-18T05:29:47.266220Z
Verification: official DOI/proceedings metadata, SOSP program, and author/institution summaries; confidence=medium.
Canonical title: Mitigating Application Resource Overload with Targeted Task Cancellation
Question: During overload, which running tasks should be cancelled to preserve SLO attainment while minimizing request loss?
Context: Front-door shedding can reject innocent new work while already-running tasks monopolize internal resources.
Method: Atropos profiles per-task resource demand and targets the tasks responsible for overload rather than cancelling indiscriminately.
Evaluation: workloads=; baselines=front-door/request shedding; metrics=SLO attainment and request loss; results=The primary sources verified the mechanism and artifact-backed SOSP publication, but accessible text did not expose unambiguous quantitative results.
Interpretation: Overload control improves when cancellation follows causal resource attribution instead of arrival order.
Reusable lesson: Attribute resource pressure to active work before choosing victims.
Applicability: Multi-stage services with SLOs and cancellable tasks during CPU/memory/resource overload.
Limits: Depends on timely attribution and safe cancellation points; quantitative evaluation remains abstract-incomplete.
자료 검증 verify_9fd4a028584f443255ad:
ev_38857c59321440ac ·
판단 보류
확인 범위: 서지정보만 확인 · 주장: context · S3ResearchAgent · 2026-07-18T14:58:44.385845Z
자료: R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=79e1cfc1cd3a2cf77a86b19613e58fa0f124fd29d9ea48597b9f30b432428a9f / 위치: 보존 파일 objects/sha256/79/79e1cfc1cd3a2cf77a86b19613e58fa0f124fd29d9ea48597b9f30b432428a9f
보존 객체는 cookie/landing page이므로 서지 위치만 확인했고 본문 주장을 검증하지 못함.
자료 검증 verify_61ebef466ccea8da488d:
verified-content-v1-0141 ·
판단 보류
확인 범위: 서지정보만 확인 · 주장: context · S3ResearchAgent · 2026-07-18T14:58:44.558851Z
자료: R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=79e1cfc1cd3a2cf77a86b19613e58fa0f124fd29d9ea48597b9f30b432428a9f / 위치: 보존 파일 objects/sha256/79/79e1cfc1cd3a2cf77a86b19613e58fa0f124fd29d9ea48597b9f30b432428a9f
보존 객체는 cookie/landing page이므로 서지 위치만 확인했고 본문 주장을 검증하지 못함.
자료 검증 verify_792ea8b52e036b0db14b:
canonical-paper-v2-90270bf8 ·
판단 보류
확인 범위: 서지정보만 확인 · 주장: context · S3ResearchAgent · 2026-07-18T14:58:44.811395Z
자료: R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=79e1cfc1cd3a2cf77a86b19613e58fa0f124fd29d9ea48597b9f30b432428a9f / 위치: 보존 파일 objects/sha256/79/79e1cfc1cd3a2cf77a86b19613e58fa0f124fd29d9ea48597b9f30b432428a9f
보존 객체는 cookie/landing page이므로 서지 위치만 확인했고 본문 주장을 검증하지 못함.
자료 검증 verify_3008d596e42eb19d68a1:
canonical-paper-v2-90270bf8 ·
반박함
확인 범위: 원문 확인 · 주장: attempt,observation,interpretation,reusable_lesson,applicability · S3ResearchAgent · 2026-07-18T14:58:44.988413Z
자료: R2-RESTIC:7f893ca5afd2cfb6fe320e9b61063ccc70e75a7a96589420038c8cf338b273be; archive-manifest-sha256=e28171fb69e141ce306d92dfe4b10e6cdc6e81d4fa910c30a846204dbcf8edf8; sha256=edc0238138cf81a0a77de826906b286c9fec89ccdf83761e103b06a457a78725; adjudicated detailed correction / 위치: DETAILED-Atropos SOSP 2025 §§1,5 and evaluation figures pre-revision source locator.
Current claims are contradicted or exceed the preserved source; revision is staged at low confidence.