Rethinking cloud abstractions for tenant-provider cooperative optimization of AI workloads

2025-01-16

AI Summary

AI workloads in multi-tenant clouds suffer from inefficient resource utilization due to bidirectional architectural and semantic misalignment between tenants and cloud providers. Method: This paper proposes the first tenant–provider co-designed cloud abstraction paradigm,重构 the cloud abstraction layer via cross-layer interface design, lightweight workload feature embedding, and verifiable collaborative policy negotiation—enabling joint optimization of task partitioning, scheduling, and fault tolerance. Contribution/Results: The work systematically identifies and models four key collaboration opportunities, establishing the first comprehensive framework for AI workload co-optimization. Experiments demonstrate significant improvements in performance, energy efficiency, and elasticity—while maintaining task stability and supporting sustainability objectives.

Technology Category

• Machine Learning: Hardware-aware ML
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

AI workloads, often hosted in multi-tenant cloud environments, require vast computational resources but suffer inefficiencies due to limited tenant-provider coordination. Tenants lack infrastructure insights, while providers lack workload details to optimize tasks like partitioning, scheduling, and fault tolerance. We propose the HarmonAIze project to redefine cloud abstractions, enabling cooperative optimization for improved performance, efficiency, resiliency, and sustainability. This paper outlines key opportunities, challenges, and a research agenda to realize this vision.

Problem

• Cloud Services
• AI Task Processing
• Resource Utilization Efficiency

Innovation

• HarmonAIze
• Collaborative Optimization
• Energy-Efficient AI Execution