Structured Evaluation of Risk-Aware Network Automation Across Heterogeneous Network and Simulated Cloud Infrastructure
This technical report presents a structured evaluation of risk-aware network automation across heterogeneous network infrastructure and simulated AWS cloud operations. Eight written change-management procedures are compared using operator-action counts, distinct tools or sessions, and a five-level ordinal scoring methodology. The network scenarios cover VLAN configuration, OSPF configuration, rollback, compliance, multi-vendor changes, risk prediction, and logical digital-twin analysis. The cloud scenario evaluates a simulated AWS security-group ingress rule. For VLAN, OSPF, and multi-vendor changes, the defined traditional procedure requires 23 operator actions, compared with 6 actions in the proposed platform procedure. These figures represent written procedures rather than measured execution times or observed device operations. The report documents implementation evidence, source-code provenance, unit-test results, scoring sensitivity, and limitations. An isolated July 2026 controller archive passed 29 of 31 tests, with two remaining failures related to stored VLAN template targets. A separate five-device VLAN configuration preview was generated, but successful five-device workflow execution and live device deployment were not demonstrated. The findings provide a reproducible framework for examining operator effort and workflow structure. They do not establish production performance improvements, live AWS mutation, or successful closed-loop recovery. Supplementary materials include scoring worksheets, test logs, configuration preview evidence, source snapshots, checksum records, and reproduction instructions.
Authors
- KARTHIK PONNAIAH (ORCID: https://orcid.org/0009-0001-2158-3060)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-05
- DOI
- https://doi.org/10.5281/zenodo.23984280
- Primary Topic
- Software-Defined Networks and 5G
- Type
- article
- Field-Weighted Citation Impact
- 0.00