Scheduling Task-parallel Applications in Dynamically Asymmetric Environments

TL;DR

This paper presents a dynamic, application-level scheduling approach that leverages moldability and task criticality to adapt to changing performance asymmetries, reducing interference in shared and distributed systems.

Contribution

It introduces a scheduler that learns platform performance characteristics and dynamically adjusts task scheduling based on criticality and parallelism.

Findings

Criticality-aware scheduling improves performance under interference.

Parallelism tuning reduces interference effects.

Dynamic learning enhances scheduling effectiveness.

Abstract

Shared resource interference is observed by applications as dynamic performance asymmetry. Prior art has developed approaches to reduce the impact of performance asymmetry mainly at the operating system and architectural levels. In this work, we study how application-level scheduling techniques can leverage moldability (i.e. flexibility to work as either single-threaded or multithreaded task) and explicit knowledge on task criticality to handle scenarios in which system performance is not only unknown but also changing over time. Our proposed task scheduler dynamically learns the performance characteristics of the underlying platform and uses this knowledge to devise better schedules aware of dynamic performance asymmetry, hence reducing the impact of interference. Our evaluation shows that both criticality-aware scheduling and parallelism tuning are effective schemes to address interference in both shared and distributed memory applications