EuroSys Reproducibility

•
1.1k
•

Hardens computer systems research papers for artifact evaluation by tracking experiment provenance, managing variance, and automating reproducible figure generation.

The EuroSys Reproducibility skill provides systems researchers with rigorous methodologies to ensure experimental claims withstand scrutiny during double-blind peer review and Artifact Evaluation Committee (AEC) evaluations. It establishes machine-readable provenance ledgers for every benchmark number, enforces statistical variance discipline across repeated runs, standardizes one-command figure regeneration pipelines, and crafts compliant data availability statements for both open-source and proprietary enterprise workloads.

Key Features

01Enforces variance discipline through multi-run spread reporting, tail latency sampling, and hardware noise mitigation
02Generates dual-phase availability statements tailored for double-blind review and camera-ready artifact badges
03Provides pre-deadline reproducibility drills to detect undocumented dependencies and numerical drift
04Structures machine-readable provenance ledgers tracking commit hashes, hardware specs, OS configurations, and seeds
05Establishes automated one-command figure regeneration pipelines from raw experimental logs
061,079 GitHub stars

Use Cases

01Auditing benchmark measurement variance and tail latency metrics across distributed testbeds
02Preparing EuroSys, OSDI, or SOSP systems research submissions for artifact evaluation badging
03Constructing anonymized data and synthetic workload fallbacks for proprietary or NDA-restricted traces