An Autonomous Deep Reinforcement Learning-based Approach for Memory Configuration in Serverless Computing

doi:10.1016/j.csi.2025.104098

An Autonomous Deep Reinforcement Learning-based Approach for Memory Configuration in Serverless Computing

2025-11-19

0

PRE

AI

M

Mostafa Ghobaei‐Arani *

R

Reza Ahsan

DOI：10.1016/j.csi.2025.104098

Original

Original request for help

Save

Abstract

En 中文

• Introducing an autonomic method using deep reinforcement learning method to predict memory configuration, and this method operates based on a reward mechanism. • Proposing an Auto Opt Mem framework, which has a MAPE (Monitor, Analyze, Plan, Execute) control loop and facilitates continuous optimization of memory allocation in a serverless environment. • Validating the effectiveness of the proposed method and demonstrate performance improvements in metrics such as latency and cost.

An Autonomous Deep Reinforcement Learning-based Approach for Memory Configuration in Serverless Computing

Abstract

Journal

Organization