Replication package for: A Systematic Literature Review of Autoscaling Techniques for Stream Processing Systems

Abstract

This dataset is the public replication package for the PRISMA 2020 systematic literature review “A Systematic Literature Review of Autoscaling Techniques for Stream Processing Systems” (2015–2025). It contains: - SLR_search_log_1.xlsx — frozen review protocol (7 July 2026), primary and supplementary search logs (executed 7 and 14 July 2026), PRISMA funnel worksheet, and the full-text extraction sheet with per-study quality scores and taxonomy coding (decision-making method, adaptation timing, elasticity dimension, deployment environment);- studies.bib — BibTeX keys S for the 139 included primary studies (S15, S68, S103, S149, S193 and S199 are commented out as EC5 exclusions);- prisma_flow.pdf — PRISMA 2020 flow diagram matching Figure 2 of the manuscript;- PRISMA_2020_checklist.pdf — completed PRISMA 2020 checklist mapping each item to a section of the manuscript;- README.md — file inventory, column definitions, and corpus notes. From 1,234 identified records, 139 primary studies are retained in the manuscript synthesis. Six records were excluded under criterion EC5 (Nos. 103 and 149: venue without verifiable peer review; Nos. 15, 68, 193 and 199: arXiv preprints without an identified peer-reviewed version). Quality assessment was applied after the inclusion/exclusion screen; no study failed the Qual/10 ≥ 6 threshold. The optional protocol enrichment string (control-theoretic and learning terms) was defined but not executed as a separate database query. Grey literature (theses, preprints, non-peer-reviewed reports) was not included. Please cite this record by the version DOI that matches n = 139. Version v1 (10.5281/zenodo.22007924) still reflects n = 143 and must be superseded by a new version on concept DOI 10.5281/zenodo.22007923.