Year: 2025
Type: preprint
Abstract: Cloud operators increasingly deploy multiple ML models in their VM allocation pipelines. In such settings, individually benign predictions can shift and compound, severely degrading performance. In a cloud provider's VM placement pipeline, CPU, memory, and lifetime prediction models jointly determine server count, live migration frequency, and network utilization; yet no existing approach can systematically stress-test how these models adversely interact. Deterministic adversarial analyzers cannot capture probabilistic ML behavior, so operators miss failures that arise only from correlated distributional shifts across models In SANJESH, we formulate a bi-level optimization that captures how the ML models behave statistically and uncovers how they adversely interact. The outer level searches over what predictions the ML models could produce under distributional uncertainty to find adversarial conditions; the inner level evaluates how the VM allocator behaves given those predictions. When we applied it to the operator's production traces, SANJESH uncovered scenarios that cause $4\times$ worse performance than the operators' evaluator detected.
Source: arXiv (Cornell University)
Authors: Bostandoost, Roozbeh, Namyar, Pooria, Kakarla, Siva Kesava Reddy, Beckett, Ryan, Segarra, Santiago, Cortez, Eli, Mallick, Ankur, Hsieh, Kevin, Fonseca, Rodrigo, Hajiesmaili, Mohammad, Arzani, Behnaz
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7111133734
Cites:
Cited by:
Related to: