Agent Skillsbrycewang-stanford/Awesome-Journal-Skills › socc-artifact-evaluation

socc-artifact-evaluation

GitHub

指导SoCC论文作者按ACM规范准备、打包和归档系统类研究的可复现工件,涵盖徽章获取、测试床描述及结果复现路径。

SoCC-Skills/skills/socc-artifact-evaluation/SKILL.md brycewang-stanford/Awesome-Journal-Skills

Trigger Scenarios

准备SoCC会议可复现工件 配置ACM Artifact Review和Badging

Install

npx skills add brycewang-stanford/Awesome-Journal-Skills --skill socc-artifact-evaluation -g -y
More Options

Non-standard path

npx skills add https://github.com/brycewang-stanford/Awesome-Journal-Skills/tree/main/SoCC-Skills/skills/socc-artifact-evaluation -g -y

Use without installing

npx skills use brycewang-stanford/Awesome-Journal-Skills@socc-artifact-evaluation

指定 Agent (Claude Code)

npx skills add brycewang-stanford/Awesome-Journal-Skills --skill socc-artifact-evaluation -a claude-code -g -y

安装 repo 全部 skill

npx skills add brycewang-stanford/Awesome-Journal-Skills --all -g -y

预览 repo 内 skill

npx skills add brycewang-stanford/Awesome-Journal-Skills --list

SKILL.md

Frontmatter
{
    "name": "socc-artifact-evaluation",
    "description": "Use when packaging an ACM SoCC artifact for the ACM Artifact Review and Badging scheme (Artifacts Available, Evaluated Functional and Reusable, Results Reproduced), covering what a cloud-systems evaluator checks first, reproducing tail-latency and cost results on a testbed, DOI-issuing archives, and the fact that whether SoCC runs a dedicated artifact-evaluation track for a given edition must be verified."
}

SoCC Artifact Evaluation

Use this for artifact preparation. SoCC — as an ACM venue — follows the ACM Artifact Review and Badging scheme where an edition offers evaluation. First, verify whether the current SoCC edition runs a dedicated artifact-evaluation track, and which badges it offers — unlike some sibling systems flagships that run a standing AE process, SoCC's artifact track and badge set are decided per edition and are 待核实 as of 2026-07-09. The advice below applies once the edition's call confirms evaluation.

Two things to internalize: badges are earned by evaluators actually reproducing your cloud results, and the review artifact (anonymized, for the paper's reviewers) is not the same deliverable as the badge artifact (de-anonymized, permanently archived).

The ACM badges (verify the current set and names)

Badge What it certifies What earns it for a cloud paper
Artifacts Available The artifact is permanently, publicly retrievable Deposit in a DOI-issuing archive (Zenodo, figshare, Software Heritage)
Artifacts Evaluated - Functional The artifact runs and does what the paper says A documented testbed setup, a workload replay, and expected outputs
Artifacts Evaluated - Reusable Others can build on it Functional plus careful docs, structure, licensing, and a portable harness
Results Reproduced An evaluator reproduced the paper's key results A turnkey path from the artifact to the headline throughput/tail/cost numbers

Available is a low-cost, high-value badge (archive the package); Functional/Reusable/Reproduced require the evaluator's own run to succeed, so the failure mode is always "did not run on their testbed," never "the idea was weak."

What a cloud-systems evaluator opens first

Claim type First thing inspected Common failure caught
A cloud system/mechanism The README and one setup+run command Undocumented cluster assumptions; only-runs-on-authors'-testbed
A measurement/trace study The scripts that turn the trace into the paper's figures Numbers in the PDF no script reproduces; trace missing
A scheduling/serverless result The workload generator + the tail/cost measurement scripts Only the mean reproduces; tail and cost cannot be regenerated
A large-scale deployment A scaled-down but faithful reproduction path Requires a proprietary cluster; no smaller-scale replay

Assume an evaluator has a bounded time budget and cannot reserve your 200-node cluster. Provide a scaled-down reproduction that still regenerates the shape of the tail and cost results, plus a clear statement of what needs full scale.

Packaging plan

[Environment] ship a Dockerfile / pinned environment AND a testbed description (node counts,
              instance types, OS, kernel) so a run is reproducible
[Workloads]   the workload generator or the (anonymized, then released) trace, not just a pointer
[README]      one-screen orientation: what it is, how to set up, how to run a small demo, how to
              reproduce each figure, expected runtime and outputs
[Mapping]     an explicit table: paper claim -> script -> expected result (incl. tail and cost)
[Scaled path] a small-scale reproduction that runs without the full cluster
[Provenance]  commit SHAs, trace extraction dates, instance types, seeds, run counts
[License]     an OSI-approved license so the artifact can be badged Reusable
[Archive]     deposit in a DOI-issuing repository for the Available badge

Anonymized review artifact vs. badge artifact

  • At submission: the artifact is anonymized for the paper's reviewers — no cluster names, provider hints, owner strings, or identity-revealing trace provenance, and no live repository that discloses authors.
  • After acceptance: replace anonymized placeholders with the public, licensed, DOI-issuing archive; this is the version any artifact evaluators badge and the camera-ready cites.

Worked vignette: packaging an autoscaler + trace study

A paper contributes a tail-aware autoscaler and a measurement of the cost-tail gap. To target Reusable and Reproduced: ship a Docker image with the controller pre-built; a run_demo.sh that replays a short trace slice on a few nodes in minutes and prints p99 and instance-seconds; a reproduce/ directory whose scripts regenerate each figure from logged runs; a claim-to-script mapping table in the README; the (released) trace-replay harness with pinned SHAs; and an MIT/Apache license. State honestly which figures are turnkey at small scale and which need the full testbed.

Calibration

  • Confirm the track exists for this edition before planning; SoCC's AE track and badge set are 待核实 per cycle.
  • Reproducing tail and cost, not just the mean, is the cloud-specific bar; design the package so an evaluator can regenerate them.
  • Badge names, the exact set offered, and whether evaluation is single- or double-anonymous vary by edition — confirm on the current call.

Output format

[Track status] SoCC AE track confirmed for this edition? yes/no/待核实
[Target badges] Available / Functional / Reusable / Reproduced
[Artifact role] anonymized review artifact / public badge artifact
[Contents] <system/workloads/trace/scripts/testbed-desc/provenance/license>
[Small-scale test] does setup + demo reproduce tail+cost shape without the full cluster? yes/no
[Claim mapping] <claim -> script -> expected result (incl. tail/cost) present? yes/no>
[Fixes before upload] <ordered list>

Version History

  • 9f86f09 Current 2026-07-19 17:41

Same Skill Collection

AAAI-Skills/skills/aaai-artifact-evaluation/SKILL.md
AAAI-Skills/skills/aaai-author-response/SKILL.md
AAAI-Skills/skills/aaai-camera-ready/SKILL.md
AAAI-Skills/skills/aaai-experiments/SKILL.md
AAAI-Skills/skills/aaai-related-work/SKILL.md
AAAI-Skills/skills/aaai-reproducibility/SKILL.md
AAAI-Skills/skills/aaai-review-process/SKILL.md
AAAI-Skills/skills/aaai-submission/SKILL.md
AAAI-Skills/skills/aaai-supplementary/SKILL.md
AAAI-Skills/skills/aaai-topic-selection/SKILL.md
AAAI-Skills/skills/aaai-workflow/SKILL.md
AAAI-Skills/skills/aaai-writing-style/SKILL.md
AAMAS-Skills/skills/aamas-artifact-evaluation/SKILL.md
AAMAS-Skills/skills/aamas-author-response/SKILL.md
AAMAS-Skills/skills/aamas-camera-ready/SKILL.md
AAMAS-Skills/skills/aamas-experiments/SKILL.md
AAMAS-Skills/skills/aamas-related-work/SKILL.md
AAMAS-Skills/skills/aamas-reproducibility/SKILL.md
AAMAS-Skills/skills/aamas-review-process/SKILL.md
AAMAS-Skills/skills/aamas-submission/SKILL.md
AAMAS-Skills/skills/aamas-supplementary/SKILL.md
AAMAS-Skills/skills/aamas-topic-selection/SKILL.md
AAMAS-Skills/skills/aamas-workflow/SKILL.md
AAMAS-Skills/skills/aamas-writing-style/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-editor-strategy/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-evidence-standards/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-literature-synthesis/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-organizing-framework/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-proposal-framing/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-review-process/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-revision/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-submission/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-tables-figures/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-topic-selection/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-workflow/SKILL.md
Academy-of-Management-Annals-Skills/skills/amann-writing-style/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-contribution-framing/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-data-analysis/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-literature-positioning/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-methods/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-rebuttal/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-review-process/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-submission/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-tables-figures/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-theory-development/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-topic-selection/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-workflow/SKILL.md
Academy-of-Management-Journal-Skills/skills/amj-writing-style/SKILL.md
Academy-of-Management-Review-Skills/skills/amr-contribution-framing/SKILL.md
Academy-of-Management-Review-Skills/skills/amr-data-analysis/SKILL.md

Metadata

Files
0
Version
5717eab
Hash
40f36f2e
Indexed
2026-07-19 17:41

- 위키
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-08-06 23:14
浙ICP备14020137号-1 $방문자$