Decouple and Decompose: Scaling Resource Allocation with DeDe

2024-12-16

AI Summary

Large-scale resource allocation in cloud systems has surpassed the scalability limits of conventional commercial solvers. To address this, we propose DeDe—a general, scalable framework that systematically uncovers the pervasive separable structure inherent in real-world resource allocation problems. DeDe introduces a novel “decoupling–decomposition” paradigm: it reformulates coupled constraints into independent resource-side and demand-side subproblems, solved via alternating optimization and parallelization, and leverages open-source convex optimization libraries (e.g., OSQP, SCS) for library-level scalability. The method provides rigorous theoretical convergence guarantees while maintaining strong engineering practicality. Evaluated on three core tasks—traffic engineering, cluster scheduling, and load balancing—DeDe achieves up to 100× speedup over state-of-the-art solvers, with allocation quality remaining within near-optimal bounds.

Technology Category

• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Distributed Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Resource allocation is fundamental for cloud systems to ensure efficient resource sharing among tenants. However, the scale of such optimization problems has outgrown the capabilities of commercial solvers traditionally employed in production. To scale up resource allocation, prior approaches either tailor solutions to specific problems or rely on assumptions tied to particular workloads. In this work, we revisit real-world resource allocation problems and uncover a common underlying structure: a vast majority of these problems are inherently separable, i.e., they optimize the aggregate utility of individual resource and demand allocations, under separate constraints for each resource and each demand. Building on this insight, we develop DeDe, a general, scalable, and theoretically grounded framework for accelerating resource allocation through a"decouple and decompose"approach. DeDe systematically decouples entangled resource and demand constraints, thereby decomposing the overall optimization into alternating per-resource and per-demand allocations, which can then be solved efficiently and in parallel. We have implemented DeDe as a library extension to an open-source solver, maintaining a familiar user interface. Experimental results across three prominent resource allocation tasks -- traffic engineering, cluster scheduling, and load balancing -- demonstrate DeDe's substantial speedups and robust allocation quality.

Problem

• Scaling resource allocation in cloud systems efficiently
• Overcoming limitations of traditional commercial solvers
• General framework for separable optimization problems

Innovation

• Decouples entangled resource and demand constraints
• Decomposes optimization into parallel per-resource allocations
• Extends open-source solver with scalable library