BitMatrix: A Multipurpose Sketch for Monitoring of Multi-tenant Networks
- PDF / 2,107,901 Bytes
- 30 Pages / 439.37 x 666.142 pts Page_size
- 0 Downloads / 181 Views
BitMatrix: A Multipurpose Sketch for Monitoring of Multi‑tenant Networks Regis Francisco Teles Martins1 · Rodolfo da Silva Villaça2 · Fábio Luciano Verdi1 Received: 6 January 2020 / Revised: 21 May 2020 / Accepted: 11 July 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Sketches are probabilistic data structures capable of summarizing and storing network data (packets, bytes, and flows), with a certain degree of accuracy, that have become widely popular for network measurement and monitoring. In this paper, we propose a new multi-purpose sketch, called BitMatrix, which is capable of working in multi-tenant networks. Owing to its multi-dimensional architecture, BitMatrix can differentiate between bit markings and byte/packet counting from different sources in a network. As a multi-purpose sketch, BitMatrix and its algorithms contribute to the literature by providing information regarding the paths traversed by each packet and are designed for use in multi-tenant networks. We also designed a statistical model to adjust the measurements owing to the probabilistic behavior of the sketches. Such a model is able to infer the standard error rate and approximate the BitMatrix counters to the real value. The adjusted BitMatrix measurement has a Mean Absolute Percentage Error of ± 6.14%. The BitMatrix sketch was implemented using P4 language and a simulator was also developed, that allowed its scaling using real traces from CAIDA in an NSF network topology. Keywords Network monitoring · Sketches · Multi-tenant networks · Programmable networks
* Rodolfo da Silva Villaça [email protected] Regis Francisco Teles Martins [email protected] Fábio Luciano Verdi [email protected] 1
Department of Computing (DCOMP‑So), Federal University of São Carlos (UFSCar), Sorocaba, SP, Brazil
2
Industrial Technology Department (DTI), Federal University of Espírito Santo (UFES), Vitoria, ES, Brazil
13
Vol.:(0123456789)
Journal of Network and Systems Management
1 Introduction Accurate network measurements are of paramount importance for monitoring, management, capacity planning, and traffic engineering purposes. In addition, cloud computing has become very popular; thus, increasing the challenges in measuring a complex network infrastructure that supports multiple services and applications. Therefore, it is becoming increasingly urgent for an accurate and fine-grained measurement system to operate efficiently when considering its use in a complex infrastructure with multiple services and tenants[1]. Faced with the need for more effective ways to measure traffic that flows in a network link between network devices, understand its behavior to support traffic engineering decisions, a valuable contribution is made through the utilization of probabilistic data structures (sketches). An accurate measurement of traffic on links and between network devices can help network managers better distribute flows into the links as a load balancing action and to find bottlenecks in the network devices that
Data Loading...
