Efficient Scaling on GPU for Federated Learning in Kubernetes: A Reinforcement Learning Approach
Authors:
- 2 July 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Abstract
Distributed learning enables efficient training with large-scale data, allowing processing to occur in multiple locations without centralization. However, the “straggler problem”, refer-ring to delayed update due to performance heterogeneity, hinders efficient learning. GPUs are crucial for enhancing learning speed by efficiently processing large-scale data through parallel computing capabilities. Efficient GPU utilization, considering resource costs, is essential for reducing training time and costs. Meanwhile, Kubernetes offers scalability, management ease, and resource provisioning for efficient operations. In this paper, we propose a Kubernetes-based vertical scaling scheme to address FL's straggler problem. For GPU scaling, we formulate an opti-mization problem that considers both resource cost and learning speed. Then, we propose a DRL-based approach to address this problem. We also leverage various Kubernetes features as well as CUDA Multi-Process Service (MPS) for execution of vertical scaling. We validate the proposed scheme's performance through various evaluations on a real testbed.Keywords
This publication has 19 references indexed in Scilit:
- Voda: A GPU Scheduling Platform for Elastic Deep Learning in Kubernetes Clusters2023
- A GPU-Accelerated Molecular Docking Workflow with Kubernetes and Apache Airflow2023
- ArkGPU: enabling applications’ high-goodput co-location execution on multitasking GPUsCCF Transactions on High Performance Computing, 2023
- FedLesScan: Mitigating Stragglers in Serverless Federated Learning2022
- Design and Implementation of Kubernetes enabled Federated Learning Platform2021
- Asynchronous Online Federated Learning for Edge Devices with Non-IID Data2020
- A Proposal of Kubernetes Scheduler Using Machine-Learning on CPU/GPU Cluster2020
- RLSK: A Job Scheduler for Federated Kubernetes Clusters based on Reinforcement Learning2020
- Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edge2019
- Gaia Scheduler: A Kubernetes-Based Scheduler Framework2018