Performance Prediction of On-NIC Network Functions with Multi-Resource Contention and Traffic Awareness

2024-05-09

Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1

AI Summary

To address multi-resource contention—particularly among on-chip accelerators and memory—induced by co-locating multiple network functions (NFs) on SmartNICs, as well as inaccurate NF performance prediction under dynamic traffic conditions, this paper proposes the first SmartNIC-aware multi-resource co-contention model integrated with fine-grained traffic awareness. We design an adaptive, high-accuracy NF performance prediction framework on the BlueField-2 platform, comprising resource behavioral modeling, traffic-sensitive feature extraction, and a lightweight regression prediction architecture. Compared to state-of-the-art methods, our approach improves prediction accuracy by 78.8% and reduces SLA violation rate by 92.2%. Moreover, it enables novel use cases such as elastic resource scheduling, thereby significantly enhancing SmartNIC programmability and service-level assurance capabilities.

Technology Category

• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Network function (NF) offloading on SmartNICs has been widely used in modern data centers, offering benefits in host resource saving and programmability. Co-running NFs on the same SmartNICs can cause performance interference due to contention of onboard resources. To meet performance SLAs while ensuring efficient resource management, operators need mechanisms to predict NF performance under such contention. However, existing solutions lack SmartNIC-specific knowledge and exhibit limited traffic awareness, leading to poor accuracy for on-NIC NFs. This paper proposes Yala, a novel performance predictive system for on-NIC NFs. Yala builds upon the key observation that co-located NFs contend for multiple resources, including onboard accelerators and the memory subsystem. It also facilitates traffic awareness according to the behaviors of individual resources to maintain accuracy as the external traffic attributes vary. Evaluation using BlueField-2 SmartNICs shows that Yala improves the prediction accuracy by 78.8% and reduces SLA violations by 92.2% compared to state-of-the-art approaches, and enables new practical usecases.

Problem

• Predict NF performance under resource contention
• Improve accuracy for on-NIC NFs
• Enhance traffic awareness in SmartNICs

Innovation

• Predicts NF performance on SmartNICs
• Considers multi-resource contention effects
• Enhances traffic awareness dynamically