On Measuring Available Capacity in High-speed Cloud Networks

📅 2025-01-31
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
Accurately measuring available bandwidth in high-speed cloud networks (≥10 Gbps) remains challenging due to insufficient hardware timestamp resolution on commodity servers. Method: This paper systematically compares four probe transmission/reception schemes and, for the first time, validates DPDK’s capability to achieve sub-microsecond hardware timestamping and rate-controllable probe generation on 10 Gbps links. We propose a dual-path measurement framework bypassing the Linux kernel: UDP-based for ≤2.5 Gbps links and DPDK-accelerated for high-throughput scenarios. Contribution/Results: The DPDK path achieves <5% measurement error on end-to-end 10 Gbps paths—significantly outperforming conventional approaches. Our framework overcomes the measurement bottleneck of general-purpose servers under high-throughput conditions, enabling lightweight, high-accuracy available bandwidth estimation. This advances traffic engineering in cloud and backbone networks by providing a practical, precision-oriented foundation for real-time capacity assessment.

Technology Category

Application Category

📝 Abstract
Measurement of available path capacity with high accuracy over high-speed links deployed in cloud and transport networks is vital for performance assessment and traffic engineering. Methods for measuring the available path capacity rely on sending and receiving time stamped probe packets. A requirement for accurate estimates of the available path capacity is the ability to generate probe packets at a desired rate and also time stamping with high precision and accuracy. This is challenging especially for measurement systems deployed using general purpose hardware. To touch upon the challenge this paper describes and evaluates four approaches for sending and receiving probe packets in high-speed networks (10+ Gbps). The evaluation shows that the baseline approach, based on the native UDP socket, is suitable for available path capacity measurements over links with capacities up to 2.5 Gbps. For higher capacities we show that an implementation based on Data Plane Development Kit (DPDK) gives good results up to 10 Gbps.
Problem

Research questions and friction points this paper is trying to address.

High-speed Cloud Network
Available Bandwidth Measurement
Network Performance Evaluation
Innovation

Methods, ideas, or system contributions that make the work stand out.

High-speed Networks
DPDK Technology
Bandwidth Measurement
🔎 Similar Papers
No similar papers found.
G
Ganapathy Raman Madanagopal
Research Area Cloud Technologies, Ericsson Research, Stockholm, Sweden
C
Christofer Flinta
Research Area Cloud Technologies, Ericsson Research, Stockholm, Sweden
Andreas Johnsson
Andreas Johnsson
Ericsson, Uppsala University
Network ManagementNetwork MeasurementMachine LearningNetwork PerformanceIoT
F
Farnaz Moradi
Research Area Cloud Technologies, Ericsson Research, Stockholm, Sweden
Daniel Turull
Daniel Turull
Ericsson