Workload Buoyancy: Keeping Apps Afloat by Identifying Shared Resource Bottlenecks

TL;DR

This paper introduces buoyancy, a new abstraction that combines application and system metrics to better identify shared resource bottlenecks in multi-tenant, heterogeneous computing environments, improving workload performance analysis.

Contribution

The paper presents buoyancy, a novel holistic metric that captures resource contention and performance headroom, enabling more effective workload orchestration in complex multi-tenant systems.

Findings

Buoyancy improves bottleneck detection accuracy by 19.3% over traditional heuristics.

It provides a holistic view of resource contention and performance headroom.

Buoyancy can replace conventional metrics to enhance system observability.

Abstract

Modern multi-tenant, hardware-heterogeneous computing environments pose significant challenges for effective workload orchestration. Simple heuristics for assessing workload performance, such as CPU utilization or application-level metrics, are often insufficient to capture the complex performance dynamics arising from resource contention and noisy-neighbor effects. In such environments, performance bottlenecks may emerge in any shared system resource, leading to unexpected and difficult-to-diagnose degradation. This paper introduces buoyancy, a novel abstraction for characterizing workload performance in multi-tenant systems. Unlike traditional approaches, buoyancy integrates application-level metrics with system-level insights of shared resource contention to provide a holistic view of performance dynamics. By explicitly capturing bottlenecks and headroom across multiple resources, buoyancy facilitates resource-aware and application-aware orchestration in a manner that is intuitive, extensible, and generalizable across heterogeneous platforms. We evaluate buoyancy using representative multi-tenant workloads to illustrate its ability to expose performance-limiting resource interactions. Buoyancy provides a 19.3% better indication of bottlenecks compared to traditional heuristics on average. We additionally show how buoyancy can act as a drop-in replacement for conventional performance metrics, enabling improved observability and more informed scheduling and optimization decisions.