Towards Resource-Efficient Compound AI Systems

TL;DR

This paper introduces Murakkab, a prototype system for resource-efficient Compound AI Systems that uses a declarative workflow model and adaptive runtime to improve efficiency and resource management.

Contribution

It proposes a novel decoupled workflow programming model and adaptive runtime system for resource-aware scheduling in Compound AI Systems, demonstrated through a prototype implementation.

Findings

Up to 3.4x faster workflow completion times

Approximately 4.5x higher energy efficiency

Effective resource optimization in AI workflows

Abstract

Compound AI Systems, integrating multiple interacting components like models, retrievers, and external tools, have emerged as essential for addressing complex AI tasks. However, current implementations suffer from inefficient resource utilization due to tight coupling between application logic and execution details, a disconnect between orchestration and resource management layers, and the perceived exclusiveness between efficiency and quality. We propose a vision for resource-efficient Compound AI Systems through a declarative workflow programming model and an adaptive runtime system for dynamic scheduling and resource-aware decision-making. Decoupling application logic from low-level details exposes levers for the runtime to flexibly configure the execution environment and resources, without compromising on quality. Enabling collaboration between the workflow orchestration and cluster manager enables higher efficiency through better scheduling and resource management. We are building a prototype system, called Murakkab, to realize this vision. Our preliminary evaluation demonstrates speedups up to $\sim 3.4\times$ in workflow completion times while delivering $\sim 4.5\times$ higher energy efficiency, showing promise in optimizing resources and advancing AI system design.