24

UDC 004.3
RESEARCH OF SDN NETWORK PERFORMANCE
PARAMETERS USING MININET NETWORK EMULATOR
Oleksandr I. Romanov, Ivan O. Saychenko, Anton I. Marinov, Serhii S. Skolets
Institute of Telecommunication Systems
Igor Sikorsky Kyiv Polytechnic Institute, Kyiv, Ukraine

Background. The implementation of the new modern services on existing networks requires replacement or
modernization of the old equipment. This negatively affects the efficiency of providing users with new types of services
and their cost. Therefore, the issues/tasks of implementing SDN technology in the construction of networks are in the
center of attention of representatives of science-research organizations and telecom operators. SDN networks have the
advantages of flexible scaling of the system without replacing existing server and network equipment, eliminating
operator dependence on specific vendor solutions, and rapid implementation of the new network technologies and
services. To determine the numerical values of indicators of the functioning of SDN networks and check their
compliance with the requirements, it is proposed to use in working process simulation modelling. For these goals, the
Open Networking Foundation consortium developed the Mininet network emulator, the elements of that are published
publicly. The formation of the simulation model and the procedure for determining the performance indicators of the
SDN network that uses Mininet has a number of features that are considered in the work.
Objective. The purpose of the article is to build a simulation model based on the Mininet network emulator and to
determine the performance indicators of SDN networks of various structures/topologies.
Methods. Creation of a SDN network segment for testing process of its functioning in the overload mode of various
communication routes.
Results. The Mininet network emulator allows simulating SDN networks of a rather complex structure, change the
performance of network branches and the amount of load in communication routes, to create a network-wide overload
mode. During the simulation, it is possible to determine a number of indicators of network performance, such as the
RTT (Round Trip Time) parameter for each route of communication, the bandwidth of branches and routes of
communication, the amount of delay on network elements, loading of OpenFlow Switch ports/interfaces, network
elements with the highest delay, the number of served and lost packets.
Conclusions. Use of the Mininet network emulator is a fairly convenient tool for determining the performance
indicators of SDN networks. However, there are also some problems. First of all, setting up the system's operability
requires writing programs to ensure the interaction among standard elements from the Mininet library. Secondone, is
when congestion occurs in the communication routes, the Floodlight Controller does not automatically balance traffic
along the work-around route. This task has to be solved manually by making changes to the OpenFlow Switch routing
tables.
Keywords: Software Defined Networking; SDN; Mininet; OpenFlow; Floodlight Controller; OpenFlowSwitch;
NFV; virtualization.

INTRODUCTION services, modification (upgrades) of equipment or

services is performed in turn in each network
Existing networks are a collection of hardware element and requires close internal and external
and software equipment such as switches, routers, resources of the operator. This approach makes the
firewalls. These devices were created on the basis of operator's network inflexible, complicates the
specific hardware and software platforms from introduction of new services and functions, and
various vendors. Therefore, the introduction of new increases the operator's dependence on vendor
modern services on existing networks, as a rule, solutions.
requires replacement or modernization of the old Therefore, the focus of representatives of
equipment staff [1,2,3,4]. This approach leads to the research organizations, universities and mobile
emergence of long design cycles, procurement of the operators is on the issues of building networks based
necessary equipment and commissioning. All this on SDN. The largest contribution to the
negatively affects the efficiency of providing users development of this product comes from the Open
with new types of services. Networking Foundation (ONF) consortium. ONF is
However, the networks of telecommunications a non-profit organization dedicated to accelerating
operators today are mainly composed of “monolithic the implementation of SDN and NVF [5].
physical” network elements, where the control, To date, the ONF has developed documents that
administration and data transfer functions are describe the principles of construction and operation
performed by physical devices. Often, a telecom of SDN networks. In works [7, 8, 9, 10] general
operator is forced to build its network using requirements, system approaches and generalized
equipment from one manufacturer, since in this case architecture of SDN networks are considered. In
it is easier to ensure the compatibility of network papers [6, 11, 12, 15], the tasks and features of the
elements and to carry out upgrades. Deployment of protocols are considered, they are used to solve

© The Author(s) 2021. Published by Igor Sikorsky Kyiv Polytechnic Institute.

This is an Open Access article distributed under the terms of the license CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/), which permits re-use,
distribution, and reproduction in any medium, provided the original work is properly cited.
O. Romanov, I. Saychenko, A. Marinov, S. Skolets. RESEARCH OF SDN NETWORK PERFORMANCE PARAMETERS 25
USING MININET NETWORK EMULATOR

various problems in SDN networks. In works [13, of the network under study. However, to date, the
14, 16] described the functions of constructing MININET platform is the most deeply developed, it
elements of the SDN network and the order of their allows you to build models of networks of varying
interaction in the process of streaming information. degrees of complexity, it is quite convenient to work
In works [17, 18, 19, 20, 21], the principles of with, and most importantly, it is in the public
constructing an optical transport network are domain.
considered and recommendations for ensuring the Let us consider the principle of constructing a
safety of their operation are given. The most SDN network model using the Mininet network
complete and systematized material is presented in emulator [5, 6]. Fig. 1 shows the structure of a data
works [22, 23]. Articles [24, 25, 26, 27] explore transmission network, which consists of standard
various aspects of information flows and focus on switches and hosts (computers). Today, these
meeting the requirements for ensuring network elements independently solve both control tasks and
security. data transmission tasks.
It should be noted that in most of the works,
special attention is paid to the implementation of
individual elements of the SDN network. However,
in the practical implementation of such a complex
system as a SDN network, a comprehensive
assessment of the performance indicators of the
system as a whole is required. For this, various types
of modelling are used to predict the behaviour of the
system in various modes of operation.

MAIN PART
To determine the numerical indicators of the
functioning of complex systems and check them
meet the requirements, three types of models are Fig. 1. Network structure emulated in Mininet
used: physical, analytical and simulation [1, 2]. While the transition to SDN, the functional
Physical models assume the deployment of a architecture of the network changes. The essence of
network section on real equipment and practical the change is that in SDN the control plane is
operation in order to determine the performance separated from the data plane. In this case, all control
characteristics. This method allows you to get values functions are transferred to the central SDN
the closest to real values. However, this approach controller, and only the executive functions remain
requires a lot of time and money. behind the data plane. Then, when switching to SDN
Analytical models are easy to use and do not principles, the functional architecture of the network
require large material investments. However, such on monolithic elements (Fig. 1) takes the form shown
complex systems as telecommunications are difficult in Fig. 2.
to describe mathematically. Usually, when
describing service processes, it is necessary to
introduce a large number of restrictions. As a result,
such models can have very low accuracy.
Therefore, as a rule, simulation models are used.
They are less expensive than physical models, allow
taking into account real network processes and have
a fairly high variability of the parameters under
study.
To study the performance of SDN networks using
the simulation method, the Open Networking
Foundation (ONF) has developed the Mininet
network emulator. Elements of Mininet are available
in the public domain on the developers' forum site [3,
Fig. 2.SDN functional architecture
4]. This allows you to use ready-made elementary
At first view, the network has become more
fragments to create simulation models of SDN
complex due to the appearance of additional
networks of various configurations. It should be
elements: a controller and functional blocks at the
noted that simulation models based on the MININET
application level. But it must be borne in mind that
platform may not work immediately. They require
the controller has taken over all the control functions.
writing matching programs and settings, the
This greatly simplified the network devices at the
complexity of which depends on the type of structure
26 INFORMATION AND TELECOMMUNICATION SCIENCES VOLUME 12 NUMBER 1 JANUARY–JUNE 2021

data plane level, and there are a lot of them on the There are two modes of work switch :
network. As for the functional blocks at the 1. Packet processing mode for data that already
application level, they have always been. They were contain information about packet in the forwarding
just not counted, as they were expected to provide tables;
additional services. A feature of functional blocks in 2. The mode of processing a packet that first
the SDN network is that they can solve certain entered the switch and for which there is no data in
control tasks at the command of the controller. the forwarding tables.
The interaction of the controller with network The first processing mode is shown in Fig. 5.
elements is provided through the "southbound"
interface using the open standard OpenFlow
protocol. Today it is the main protocol recommended
by the ONF for use in SDN networks.
An important advantage of SDN networks is that
both hardware monolithic devices and virtual ones
can be used as network elements. Fig. 3 shows an
example where virtual vSwitch (for example, located
in the Data Processing Center) are connected to the
hardware WAN switches of the network (for
example, the Inernet).

Fig. 5. Mode of processing packet for which all

data contains in the Flow table.
The essence of the processing process is that a
packet from the H1 host enters port 1 and then is
transmitted for analysis to the Flow table. Checking
the set of packet fields will allow you to determine
Fig. 3.Building an SDN network using hardware and the action - to transmit the packet to port 3, to which
virtual network elements the H3 host is connected.
The processing mode of a packet that first entered
Each switch, hardware Switch or virtual vSwitch, the switch and for which there is no data in the
contains one or more flow tables, a group table, and forwarding tables is shown in Fig. 6.
supports an OpenFlow channel to communicate with
the SDN Controller (Fig. 4). Each flow table in
OpenFlow Switch contains a set of flow entries and
rules.

Fig. 4. Functional diagram of OpenFlow Switch

In Fig. 4 shows that a switch can use multiple
forwarding tables. Tables are numbered starting from
0. The packet is sequentially transmitted for analysis
across all tables. Once a package meets the selection
criteria, a specific command is executed, such as a
redirect. The forwarding entry indicates the port,
which can be physical or virtual. There are special Fig. 6. Packet processing mode for which there is
virtual ports that can change the logic of packet no data in the Flow table
processing, for example, to provide that it is
transmitted through all ports of the OpenFlow The essence of the processing process is that a
switch. packet from the H1 host enters port 1 and then is
O. Romanov, I. Saychenko, A. Marinov, S. Skolets. RESEARCH OF SDN NETWORK PERFORMANCE PARAMETERS 27
USING MININET NETWORK EMULATOR

transmitted for analysis to the Flow table. Checking structures from Mininet to hardware switches;
the set of packet fields will not determine the port to  deploying a network in Mininet on a laptop,
which the packet is to be sent. Then, as a necessary server, virtual machine or cloud.
action, the packet will be transferred to the There are three ways to build a simulation model
OpenFlow agent, which will establish a secure using Mininet:
communication channel with the SDN Controller and 1. Simultaneously installation of the network
transfer the packet to it for analysis. SDN Controller structure and controller in one file. In this case,
will determine the destination port of the packet, the controller has an embedded network.
make changes to the Flow table, and all subsequent 2. Separate installation of the network
packets of this type will be processed in the first structure and controller in different files. In this
mode. The packet processing algorithm in this mode case, we have a controller, which is a remote
is shown in Fig. 7. device that is not built into the network and is
launched when the network file is started.
3. Combined installation:
- established network structure with built-in
controller;
- an additional Floodlight controller is
installed and runs separately from the network.
The first method allows you to build only the
simplest network structures. The second method
provides the construction of more complex
structures, however, it has limitations on the modes
of operation of the network under study. The third
method has more possibilities than the first two. It
Fig. 7. Algorithm for processing a package for allows you to build large-scale network topologies.
which there is no data in the Flowtable. Floodlight controller is widely used in research in
educational institutions and is actively developing
Let's make a simulation of the network, the [7, 8]. Therefore, the work used the third method for
structure of which is shown in Fig. 8, using the constructing the Mininet simulation model.
Mininet network emulator. Floodlight controller was developed by an open
community of developers who support the
OpenFlow protocol. Modelling SDNs on the Mininet
platform using the Floodlight controller has next set
of advantages [1]:
- controller makes it easy to adapt and develop
applications that are written in Java;
- there are included REST APIs, that simplify
interface programming;
- the Floodlight website provides coding
examples to help developers create own product;
- controller works with both physical and virtual
switches which compatible with OpenFlow;
- controller can support networks on traditional
switches which does not support OpenFlow.
Fig. 8.The structure of the investigated network The structure and main functional elements of the
Mininet Floodlight controller are shown in Fig. 9.

Using Mininet, it is possible to explore SDN

networks of various structures. The main advantages
of Mininet are:
 fast network creation using standard
components.
 the ability to build networks of complex
structure;
 creation of virtual network elements such as
OpenFlow Switch, web servers, monitoring
tools, Wireshark;
 transfer and installation of configured network
28 INFORMATION AND TELECOMMUNICATION SCIENCES VOLUME 12 NUMBER 1 JANUARY–JUNE 2021

- UDP;
- TCP.
3. Amount of the load in the paths of
communication.
4. Set performance of network branches.
5. Packet processing time in network elements.
As a result of modelling the process of network
work, were determined the following indicators:
1. Network bandwidth depends on the size of the
input load to the network.
2. Total real network branch performance.
3. Load on the interfaces of each network switch.
4. RRT for all communication routes for two types
of load:
- RTT for TCP packets;
- RTT for UDP packets;
RTT represents the sum of the time it takes for
delivery the package to the consumer and the time it
takes to get confirmation that the packet has been
received.
5. Impact of latency in network elements on
Fig. 9. Functional structure of the Floodlight bandwidth performance.
controller To generate traffic in the network, were used
Now let’s consider the purpose of the basic Iperf tool to generate packets over TCP and Iperfudp
elements. The main component in the work of to generate packets over UDP.
Floodlight Controller is REST Applications – The results of network modelling in Fig. 8 are
programs that use services which use REST API, presented in Table 1.
controlled by controller modules and modular Table 1.Networkperformanceindicators
applications [2, 3].
REST Applications in their turn interact with Transmitted
Network Latency
Circuit Pusher which from REST API provides Load Bandwidth and lost
route (ms)
management of packets of flows on all switches on packets
the basis of IP-addresses with the set priority. Also 6 18.621 15201/5238
H1-H5 3.7 Mbps
Mbps ms (34% loss)
REST Applications interact with modular
4 1.787 10178/3741
applications that acting like the applications with H3-H5
Mbps
2.4 Mbps
ms (37% loss)
higher bandwidth for communication with the 5 1.683 12674/4847
controller. H4-H5 2.97 Mbps
Mbps ms (38% loss)
Modular applications include applications such During the research, Host 5 was used as a server
as: Firewall, Forwarding, Concentrator (Hub), that receives traffic from three clients - Host 1, Host
Academic Switch, Entry Pusher, Virtual Network 3, Host 4. At the initial setup, the performance of all
Filter, PortDown Reconciliation. The interaction of channels in the network was 10 Mbit / s. Further, the
all these elements creates the controller’s main load in the network for each client increased until
functionality. the network was overloaded.
In the research, using the Mininet simulation For example, a branch of the network has a
model with the Floodlight Controller, were performance of 10 Mbps. The load in the
investigated the following performance indicators of communication route was increased for reaching the
the SDN network, which is shown in Fig. 8: total incoming load in 15 Mbps. The goal was for
1. Network performance; traffic to be served not only along the shortest path,
2. Bandwidth of network channels during but also along the roundabout path. That is, there
transmitting UDP and TCP traffic packets; was a need to balance traffic. In our case, to prevent
3. Possibility of balancing traffic by the Floodlight the occurrence of packet loss in the direction from
controller in the overload mode of network paths; sw-1 to sw-2, some of the sent packets were serviced
4. RTT Packets delay values. in the transmission direction from sw-1 to sw-4. The
To discovery these parameters, there are required obtained results are shown in Table 1.The
following input data: visualization of the overloading process is shown in
1. Network structure that can be specified as a Fig. 10.
connectivity matrix;
2. Data transfer protocols that are used:
O. Romanov, I. Saychenko, A. Marinov, S. Skolets. RESEARCH OF SDN NETWORK PERFORMANCE PARAMETERS 29
USING MININET NETWORK EMULATOR

The pingall tool is responsible for determining

the RTT parameter. But to get more detailed
statistics use the command: pingallfull.
The results of the RTT research are shown in
Table2.
Table 2.Theresultsofthe RTT research
Average RTT, Number of working
Delay, ms
ms routes
0 0.146 30
2.5 22.266 30
5 38.578 30
50 308.414 30
100 909.721 30
200 1792.005 30
400 3358.825 28
450 3693.478 27
Fig.10.Network congestion (overload) 500 3131.509 18

The Wireshark utility allows analysing the Visualization of the obtained results is presented
amount of traffic that went through the in Fig.13 and Fig. 14.
corresponding ports/interfaces. Fig. 11 shows an
example of loading the eth4 interface of the switch
sw-1 during traffic transmission to Host 5, which
belongs to the switch sw-2.

Fig. 13. Dependence of the number of working and

denied communication routes with different delays
in the network

Fig. 11 Load on the eth4 interface of the switch

sw-1
The total amount of transmitted and lost packets is
shown in Fig. 12.

Fig. 14. Dependence of RTT change at different

network delays
As can be seen from Fig. 14, when packets with a
set delay of 400 ms are transmitted in the network,
the first disconnections of the routes begin to occur.
Namely, 2 directions with the largest number of
hops, with 30 directions in the network (where hop
Fig. 12 Packet transfer statistics between clients is the distance between two nodes in the network, so
and server the more hops – the more complicated the routing
Next, the value of the delay of RTT packets was path and the further are the nodes from each other).
researched [4]. According to Fig. 14, you can see a decrease in
RTT with a delay of 450 ms, due to the failure of 3
30 INFORMATION AND TELECOMMUNICATION SCIENCES VOLUME 12 NUMBER 1 JANUARY–JUNE 2021

routes. With a delay of 500 ms, we see a much larger 10. IETF RFC 7426. Request for Comments:
number of rejections, namely 12 routes. 7426. ISSN: 2070-1721 EICT. Category:
Also, as a result of the experiment during seting Informational K. Pentikousis, Ed.,
up all the above delays (Table 2), we noticed rutes 11. ISSN: 2070-1721 EICT
with the largest delays: h1-h4, h1-h5, h1-h6, h2-h4, 12. https://tools.ietf.org/html/rfc7426
h2-h5, h2- h6, h3-h4, h3-h5, h3-h6, h4-h5, h4-h6, 13. ONF TR-525 SDN Interoperability Event
h5-h4, h5-h6, h6-h4. Technical Issues Report AppFest 2015
Thus, the use of Mininet allows you to model a 14. Testing-Interop Working Group Version 0.2
fairly complex SDN network and get important 15. O.I. Romanov; M. V. Oryschuk; Y. S.
performance indicators, such as bandwidth, quality Hordashnyk. Computing of influence of stimulated
of service, packet delay using different protocols, Raman scattering in DWDM telecommunication
such as TCP, UDP and others. Mininet is publicly systems. 2016 International Conference Radio
available. Therefore, it deserves a more detailed Electronics & Info Communications (UkrMiCo)
study. Year:2016.DOI:10.1109/UkrMiCo.2016.7739622,eI
REFERENCES SBN:978-1-5090-4409-2
1. Romanov O.I., Nesterenko M.M. 16. Globa L., Skulysh M., Romanov O.,
Telecommunication network technologies Nesterenko M. (2019) Quality Control for Mobile
Evolution//Advances in the telecommunications 2019/ Communication Management Services in Hybrid
According to scientific editorship IlchenkoM.Yu., Environment. In: Ilchenko M., Uryvsky L., Globa
Kravchuk S.O., Kyiv, 2019, - p.p. 88-107.
L. (eds) Advances in Information and
2. Oleksandr I. Romanov, Mykola M. Nesterenko,
Nadiia O. Fesokha, Volodymyr B. Mankivskyi. Communication Technologies. UKRMICO 2018.
Evaluation of productivity virtualization technologies of Lecture Notes in Electrical Engineering, vol 560.
switching equipment telecommunications networks. Springer, Cham. DOI https://doi.org/10.1007/978-
Information and Telecommunication Sciences, 2020, 3-030-16770-7_4, Print ISBN978-3-030-16769-1
Volume 11, Number 1 (20), page 53 – 58, DOI: 17. ONF TR-539 OpenFlow Controller
https://doi.org/10.20535/2411-2976.12020.53-58. Benchmarking Methodologies Version 1.0
3. Romanov A.., Nesterenko М., Маnkivskyi V, November 2016
Saichenko, I. (2020). 18. ONF TS-029. MPLS-TP OpenFlow Protocol
Modelofoptimalloaddistributioninmobileoperatoracc Extensions for SPTN. Version 1.0 June 16, 2017
essnetwork.BulletinofCherkasyStateTechnologicalU 19. ONF TR-539. OpenFlow Controller
niversity, (3), 20–29. https://doi.org/10.24025/2306- Benchmarking Methodologies Version 1.0
4412.3.2020.200346 November 2016
4. Oleksandr I. Romanov, Mykola M. 20. ONOS. Security and Performance. Analysis:
Nesterenko, Volodymyr B. Mankivskyi. Used Report No. 1. Date: September 19, 2017
regress model coefficient utilization of channels for 21. ONOS Security and Performance Analysis
plan load distribution in network. Visnyk NTUU (Report No. 2). Date: Nov. 2, 2018
KPI Seriia – 22. ONF TR-537. Negotiable Datapath Model
RadioteknikaRadioaparatobuduvannia№67, and Table Type Pattern Signing Version 1.0. ONF
2016p.p. 34-42 TR-537 2016-09-08
5. Open Network Foundation. Accelerating the 23. Oleksander
Adoption of SDN & NFV. Romanov,VolodymyrMankivskyi. Optimal Traffic
http://opennetworking.wpengine.com/wp- Distribution Based on the Sectoral Model of
content/uploads/2013/03/ONF%20Overview%204- Loading Network Elements /2019 IEEE
pager%20FINAL.pdf, 2021 International Scientific-Practical Conference
6. ONF. OpenFlow Switch Specification Problems of Infocommunications, Science and
Version 1.5.1 ( Protocol version 0x06 ), Technology (PIC S&T), 8-11 Oct. 2019, Date
2014.https://opennetworking.org/wp- Added to IEEE Xplore: 09 April 2020, Conference
content/uploads/2014/10/openflow-switch- Kyiv,
v1.5.1.pdf Ukraine.DOI:10.1109/PICST47496.2019.9061296,
7. ONF TR-502. SDN architecture. Issue 1 eISBN: 978-1-7281-4184-8.
June, 2014 24. Romanov, O., Nesterenko, M., Veres, L.,
8. ONF TR-508. Requirements Analysis for Kamarali, R., Saychenko, I. Methods for calculating
Transport OpenFlow/SDN. V1.0 August 20, 2014. the performance indicators of ip multimedia
ONF TR-508 subsystem (IMS). Lecture Notes in Networks and
9. ONF specification. OpenFlow Table Type Systems, 2021, 152, с. 229-256
Patterns. Version No. 1.0 15 August 2014 25. Next-Gen SDN Tutorial, 2019,
https://opennetworking.org/wp-
O. Romanov, I. Saychenko, A. Marinov, S. Skolets. RESEARCH OF SDN NETWORK PERFORMANCE PARAMETERS 31
USING MININET NETWORK EMULATOR

content/uploads/2019/10/NG-SDN-Tutorial- and Data Plane Communication with IBC. Hindawi

Overview.pdf Publishing Corporation Mobile Information Systems
26. Carmelo CasconeO'Connor, Thomas Volume 2016, Article ID 1708970, 12 pages
Vachuska, Bruce Davie. Software-Defined http://dx.doi.org/10.1155/2016/1708970.
Networks: A Systems Approach, – 2021, р.152 29. DouglasComer, AdibRastegarnia.
27. K´evinPhemius, Mathieu Bouet and Externalization of Packet Processing in Software
J´er´emieLeguay. DISCO: Distributed Multi-domain Defined Networking. All content following this page
SDN Controllers. Thales Communications & was uploaded by Adib Rastegarniaon 11 January
Security. arXiv:1308.6138v2 [cs.NI] 29 Aug 2013 2019.
28. JunHuyLam,Sang-GonLee,Hoon-Jae Lee, https://www.researchgate.net/publication/330276437
and YustusEkoOktian. Securing SDN Southbound

Романов О.І., Сайченко І.О., Марінов А.І., Сколець С.С.

Дослідження параметрів продуктивності мережі SDN з використанням мережевого емулятора
MININET
Проблематика. Принципи побудови і функціонування мереж SDN суттєво відрізняються від традиційних
телекомунікаційних мереж. Для визначення чисельних значень показників функціонування такого типу мереж
та перевірки їх відповідності вимогам, консорціумом OpenNetworkingFoundation розроблений мережевий
емулятор Mininet. Елементи мережевого емулятора є у відкритому доступі. Однак, побудова імітаційної моделі
і порядок визначення показників функціонування мережі SDN з використанням Mininet має ряд
особливостей,які розглянуті в роботі.
Мета дослідження. Створення імітаційної моделі на базі мережевого емулятору Mininet і визначення
показників функціонування мереж SDN різної структури.
Методика реалізації. Визначення показників функціонування мережі SDN у режимі перевантаження
напрямків зв'язку.
Результати дослідження. Проведено моделювання мереж SDN складної структури і визначені наступні
показники функціонування: RTT (RoundTripTime) для кожного напрямку зв'язку, пропускна здатність гілок і
напрямків зв'язку, величина затримки на мережевих елементах, завантаження портів OpenFlowSwitch, елементи
мережі з найбільшою затримкою, число обслугованих і втрачених пакетів.
Висновки. Використання мережевого емулятора Mininet є досить зручним інструментом для визначення
показників функціонування мереж SDN. Однак, є і проблеми. По-перше, налаштування працездатності системи
вимагає написання програм для забезпечення взаємодії стандартних елементів з бібліотеками Mininet. По-друге,
при виникненні перевантажень в напрямках зв'язку, FloodlightController не забезпечує автоматичного
балансування трафіку по обхідним шляхам. Цю задачу доводиться вирішувати в ручному режимі шляхом
внесення змін до таблиці маршрутизаціїOpenFlowSwitch.
Ключові слова: Програмно-конфігуровані мережі; SDN; Mininet; OpenFlow; Floodlight контролер; OpenFlow
комутатор; NFV; віртуалізація.

Романов А.И., Сайченко И.О., Маринов А.И., Сколец С.С.

Исследование параметров производительности сети SDN с использованием сетевого эмулятора
MININET
Проблематика. Принципы построения и функционирования сетей SDN существенно отличаются от
традиционных телекоммуникационных сетей. Для определения численных значений показателей
функционирования такого типа сетей и проверки их соответствия требованиям, консорциумом Open Networking
Foundation разработан сетевой эмулятор Mininet. Элементы сетевого эмулятора имеются в открытом доступе.
Однако, построение имитационной модели и порядок определения показателей функционирования сети SDN с
использованием Mininet имеет ряд особенностей, которые рассмотрены в работе.
Цель исследования. Создание имитационной модели на базе сетевого эмулятор Mininet и определения
показателей функционирования сетей SDN различной структуры.
Методика реализации. Определение показателей функционирования сети SDN в режиме перегрузки
направлений связи.
Результаты исследования. Проведено моделирование сетей SDN сложной структуры и определены
следующие показатели функционирования: RTT (RoundTrip Time) для каждого направления связи, пропускная
способность ветвей и направлений связи, величина задержки на сетевых элементах, загрузка портов
OpenFlowSwitch, элементы сети с наибольшей задержкой, число обслуженных и потерянных пакетов.
32 INFORMATION AND TELECOMMUNICATION SCIENCES VOLUME 12 NUMBER 1 JANUARY–JUNE 2021

Выводы. Использование сетевого эмулятора Mininetявляется достаточно удобным инструментом для

определения показателей функционирования сетей SDN. Однако, есть и проблемы. Во-первых, настройка
работоспособности системы требует написания программ для обеспечения взаимодействия стандартных
элементов с библиотеками Mininet. Во-вторых, при возникновении перегрузок в направлениях связи,
FloodlightController не обеспечивает автоматической балансировки трафика по обходным путям. Эту задачу
приходится решать в ручном режиме путем внесения изменений в таблицы маршрутизации OpenFlowSwitch.
Ключевые слова: Программно-конфигурируемые сети; SDN; Mininet; OpenFlow; Floodlight контроллер;
OpenFlow коммутатор; NFV; виртуализация.