Efficient and Reuseable Cloud Configuration Search Using Discovery Spaces

2025-06-26

AI Summary

In cloud environments, efficiently searching for cost-optimal resource configurations that satisfy SLA constraints is challenging due to the high-dimensional, heterogeneous nature of configuration spaces across diverse workloads (e.g., LLM inference, big data analytics). Method: This paper proposes Discovery Space—a novel abstraction framework that unifies cross-workload configuration exploration as a structured, reusable, and distributed search space. It enables safe and transparent knowledge transfer among optimizers via formal modeling of search spaces, distributed metadata coordination, incremental knowledge distillation, and alignment of semantically similar search subspaces. Contribution/Results: Experiments demonstrate over 90% improvement in search efficiency. The framework exhibits strong generalization and reusability across multiple real-world workloads, significantly reducing cloud deployment costs and SLA violation risks while maintaining rigorous performance guarantees.

Technology Category

• Search and Optimization: Distributed Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Machine Learning: Optimization

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Finding the optimal set of cloud resources to deploy a given workload at minimal cost while meeting a defined service level agreement is an active area of research. Combining tens of parameters applicable across a large selection of compute, storage, and services offered by cloud providers with similar numbers of application-specific parameters leads to configuration spaces with millions of deployment options. In this paper, we propose Discovery Space, an abstraction that formalizes the description of workload configuration problems, and exhibits a set of characteristics required for structured, robust and distributed investigations of large search spaces. We describe a concrete implementation of the Discovery Space abstraction and show that it is generalizable across a diverse set of workloads such as Large Language Model inference and Big Data Analytics. We demonstrate that our approach enables safe, transparent sharing of data between executions of best-of-breed optimizers increasing the efficiency of optimal configuration detection in large search spaces. We also demonstrate how Discovery Spaces enable transfer and reuse of knowledge across similar search spaces, enabling configuration search speed-ups of over 90%.

Problem

• Finding optimal cloud resources for cost-effective workload deployment
• Managing large configuration spaces with millions of deployment options
• Enabling knowledge reuse for faster configuration search

Innovation

• Discovery Space abstraction for workload configuration
• Safe transparent data sharing between optimizers
• Knowledge transfer across similar search spaces