Abstract:Modern data center power distribution networks place greater demands on the stability and reliability of power supply. Growing network computing demands and complex network environments can cause network congestion, which in turn leads to network traffic overload and power supply equipment overload. Therefore, network congestion is one of the most important problems faced by data center power distribution networks. In this paper, we propose an approach called ACC-RL based on reinforcement learning (RL), which can effectively avoid network congestion and improve energy performance. ACC-RL models the congestion control task as a Partially Observable Markov Decision Process (POMDP). It is independent of the estimated value function and supports deterministic policies. It also sets the reward value function using real-time network information such as the transmission rate, RTT, and switch queue length, with the target transmission rate as the target equilibrium point. ACC-RL is highly general, can be trained on datasets running in different network environments, and generates a robust congestion control policy. The experimental results show that ACC-RL can solve the congestion problem without any predefined scenarios in different network environments. It can control the network traffic well, thus ensuring the stability and reliability of the power supply in the distribution network. We conduct network simulation experiments through NS-3. We set up different scenarios for experiments and data analysis in many-to-one, all-to-all, and long–short network environments. Compared with the popular rule-based congestion control algorithms such as TIMELY, DCQCN, and HPCC, ACC-RL shows different degrees of energy performance advantages in network metrics such as fairness, link utilization, and throughput.

A Novel Ad-hoc Network Congestion Forecast Method Based on Fuzzy Neural Network

An Active Queue Management Scheme Using Neural Network Based Predictive Control

Congestion Control for ATM Multiplexers Using Neural Networks: Multiple Sources/single Buffer Scenario

Congestion Avoidance, Detection and Alleviation in Wireless Sensor Networks

An Efficient Fuzzy based Congestion Control Technique for Wireless Sensor Networks

Congestion Control for High Dynamic Ad Hoc Networks: Price Cooperation and Receding Optimization

MPCC: Optimal Congestion Control for Multipath Routing in Ad Hoc Networks

Optimal Congestion Control Algorithm for Ad Hoc Networks: Penalty Function-Based Approach

Neural network based feedback scheduler for networked control system with flexible workload

A Short-term Traffic Flow Forecasting Method Based on Combination of Fuzzy C-mean Clustering and Neural Network

Network congestion control based on neural network prediction algorithm

A Novel Congestion Control Scheme using Firefly Algorithm Optimized Fuzzy-PID Controller in Wireless Sensor Network

Adaptive Congestion Prediction in Vehicular Ad-hoc Networks (VANET) Using Type-2 Fuzzy Model to Establish Reliable Routes

A Novel ECN-Based Congestion Control and Avoidance Algorithm with Forecasting and Verifying

The Research Of Active Network Congestion Control Algorithm

ACC-RL: Adaptive Congestion Control Based on Reinforcement Learning in Power Distribution Networks with Data Centers

A Novel Predictive Congestion Control Algorithm for Urban Traffic Network

The Research of Active Network Congestion Control Algorithm Based on Operational Data

A Congestion Control Algorithm for Ad Hoc Network

An Intelligent Call Admission Control Decision Mechanism for Wireless Networks

A Neural Network Adaptive Controller for Explicit Congestion Control with Time Delay