Inference Load-Aware Orchestration for Hierarchical Federated Learning

TL;DR

This paper introduces an inference load-aware orchestration scheme for hierarchical federated learning that optimizes model placement and device association to reduce inference latency and communication costs in continual learning scenarios.

Contribution

It presents a novel joint orchestration approach that considers inference workloads and processing capacities to improve efficiency in hierarchical federated learning.

Findings

Significant inference latency reduction achieved.

Communication costs are drastically reduced.

Optimized aggregator placement improves performance.

Abstract

Hierarchical federated learning (HFL) designs introduce intermediate aggregator nodes between clients and the global federated learning server in order to reduce communication costs and distribute server load. One side effect is that machine learning model replication at scale comes "for free" as part of the HFL process: model replicas are hosted at the client end, intermediate nodes, and the global server level and are readily available for serving inference requests. This creates opportunities for efficient model serving but simultaneously couples the training and serving processes and calls for their joint orchestration. This is particularly important for continual learning, where serving a model while (re)training it periodically, upon specific triggers, or continuously, takes place over shared infrastructure spanning the computing continuum. Consequently, training and inference workloads can interfere with detrimental effects on performance. To address this issue, we propose an inference load-aware HFL orchestration scheme, which makes informed decisions on HFL configuration, considering knowledge about inference workloads and the respective processing capacity. Applying our scheme to a continual learning use case in the transportation domain, we demonstrate that by optimizing aggregator node placement and device-aggregator association, significant inference latency savings can be achieved while communication costs are drastically reduced compared to flat centralized federated learning.