CARA: Exact Local Repair with Fresh One-Action Certification for Cloud Consolidation

2026-07-31

AI Summary

This work addresses critical limitations in existing cloud resource scheduling repair methods, which suffer from invalidated reliability verification due to reused search scenarios and compromised weak objectives in multi-objective optimization. To overcome these issues, the authors propose Certificate-Aligned Recomposition (CARA), a novel approach that decouples adaptive action generation from deployment decisions. CARA precisely recovers the top-P diverse actions within a two-host neighborhood and incorporates an independent fresh-pair certification mechanism to ensure simultaneous improvement across four key dimensions. For the first time, CARA integrates auditable recovery of local repair actions with robust certification. Evaluated across 128 synthetic environments, the resulting policies consistently outperform baselines on all five metrics: achieving a 3.57 percentage-point reduction in the J metric, 21%–28% lower continuous load, an average absolute ranking sensitivity gap below 5.3×10⁻⁴, and a false declaration probability no greater than 5%.

Abstract

Simulator-based placement pipelines may inspect many repairs but deploy only when several reliability criteria improve together. Reusing search scenes to test the selected action invalidates nominal evidence, while scalarization can trade away the weakest criterion. We introduce Certificate-Aligned Recomposition (CARA), an incumbent-anchored pipeline that separates adaptive proposal generation from a one-use deployment decision. In a bounded two-host neighborhood, a packing-specific admissible bound recovers the exact top-$P$ distinct actions under a mixed paired-binary/fixed-bet certificate order. Held-out views then select one action and commit its betting plans before fresh paired Certification opens. Under the stated sign-independence and conditional-moment assumptions, the probability of falsely declaring four-way improvement is at most $.05$, irrespective of the size or complexity of upstream search. In a prospectively frozen study over 128 independent synthetic environments and four repeated contexts, the complete fail-closed terminal policy improved a development-selected same-host-count incumbent on all five Evaluation endpoints in every environment. $J$ fell by 3.57 percentage points and the continuous burdens by 21--28%. A matched ordering sensitivity produced absolute mean gaps below $5.3\times10^{-4}$ and does not establish order superiority. CARA thus couples exact auditable proposal recovery to selection-robust fresh certification for last-mile local repair.

Problem

• cloud consolidation
• local repair
• reliability certification
• deployment decision
• multi-criteria improvement

Innovation

• Certificate-Aligned Recomposition
• exact local repair
• one-action certification
• cloud consolidation
• selection-robust validation