Treffer: The Kubernetes Network Driver Model: A Composable Architecture for High-Performance Networking

Title:
The Kubernetes Network Driver Model: A Composable Architecture for High-Performance Networking
Authors:
Ojea, Antonio
Source:
2025 IEEE 50th Conference on Local Computer Networks (LCN). :1-7
Publication Status:
Preprint
Publisher Information:
IEEE, 2025.
Publication Year:
2025
Subject Terms:
Networking and Internet Architecture (cs.NI), FOS: Computer and information sciences, Artificial Intelligence (cs.AI), Networking and Internet Architecture, Artificial Intelligence
Document Type:
Fachzeitschrift Article
DOI:
10.1109/lcn65610.2025.11146291
DOI:
10.48550/arxiv.2506.23628
Access URL:
http://arxiv.org/abs/2506.23628
Rights:
STM Policy #29
arXiv Non-Exclusive Distribution
Accession Number:
edsair.doi.dedup.....e7a4fcc4c54a7834ca8683a17d6979c6
Database:
OpenAIRE

Weitere Informationen

Traditional Kubernetes networking struggles to meet the escalating demands of AI/ML and evolving Telco infrastructure. This paper introduces Kubernetes Network Drivers (KNDs), a transformative, modular, and declarative architecture designed to overcome current imperative provisioning and API limitations. KNDs integrate network resource management into Kubernetes' core by utilizing Dynamic Resource Allocation (DRA), Node Resource Interface (NRI) improvements, and upcoming OCI Runtime Specification changes. Our DraNet implementation demonstrates declarative attachment of network interfaces, including Remote Direct Memory Access (RDMA) devices, significantly boosting high-performance AI/ML workloads. This capability enables sophisticated cloud-native applications and lays crucial groundwork for future Telco solutions, fostering a "galaxy" of specialized KNDs for enhanced application delivery and reduced operational complexity.
6 pages, 9 figures, submitted to IEEE LCN Special Track on Cloud-AI-Native Mobile Networks Powered by eBPF (CAMe 2025)