Combined power management and congestion control in High-Speed Ethernet-based Networks for Supercomputers and Data Centers
🤖 AI Summary
Addressing the mutual constraint between congestion control and energy efficiency in high-performance computing and data center high-speed Ethernet, this paper proposes a joint optimization framework integrating power management and congestion control. We model their coupling under dynamic traffic to reveal the performance–energy trade-off mechanism; further, we integrate traffic scheduling, link-rate adaptation, and fine-grained power modeling to jointly optimize congestion mitigation and idle-power suppression. Experiments demonstrate that our approach significantly improves throughput and fairness under high load, reduces idle power consumption by up to 42%, increases network resource utilization by 31%, and enhances system scalability. The core contribution is the first formulation of a congestion–power co-control paradigm specifically for high-speed Ethernet—providing both theoretically grounded and practically deployable solutions for green high-performance networking.
Application Category
📝 Abstract
The demand for computer in our daily lives has led to the proliferation of Datacenters that power indispensable many services. On the other hand, computing has become essential for some research for various scientific fields, that require Supercomputers with vast computing capabilities to produce results in reasonable time. The scale and complexity of these systems, compared to our day-to-day devices, are like comparing a cell to a living organism. To make them work properly, we need state-of-the-art technology and engineering, not just raw resources. Interconnecting the different computer nodes that make up a whole is a delicate task, as it can become the bottleneck for the whole infrastructure. In this work, we explore two aspects of the network: how to prevent degradation under heavy use with congestion control, and how to save energy when idle with power management; and how the two may interact.
Problem
Research questions and friction points this paper is trying to address.
Managing congestion control in high-speed Ethernet networks
Implementing power management for energy efficiency during idle periods
Investigating interactions between congestion control and power management
Innovation
Methods, ideas, or system contributions that make the work stand out.
Combined power management and congestion control
Prevents network degradation under heavy use
Saves energy during idle network periods
🔎 Similar Papers
No similar papers found.
