Pengembangan Aplikasi Pemantauan Jaringan Berbasis Web pada Software-Defined Networking dengan Protokol SFLOW

<p><em>Software-defined networking</em> (SDN) merupakan salah satu arsitektur jaringan yang dapat diprogram untuk memudahkan manajemen jaringan menggunakan aplikasi pengontrol.. SDN <em>controller</em> seperti ONOS hanya berfungsi untuk manajemen <em>flow</em> <em>table</em> dan tidak memiliki fitur pemantauan <em>traffic</em> jaringan yang cukup untuk mendukung proses manajemen jaringan. Oleh karena itu, untuk melakukan pemantauan <em>traffic</em> SDN maka diperlukan protokol lain seperti sFlow. Pada penelitian ini dikembangkan aplikasi sistem pemantauan <em>traffic</em> SDN berbasis web dengan menggunakan ONOS sebagai SDN <em>controller</em>, sFlow-RT sebagai sFlow <em>collector, </em>dan Node.js sebagai web <em>server</em>. Hasil pengembangan aplikasi menghasilkan tiga buah fitur utama yaitu topologi, grafik <em>traffic</em>, dan laporan. Visualisasi topologi dibuat berdasarkan data topologi dari API ONOS dan ditampilkan menggunakan pustaka vis.js. Kemudian untuk grafik <em>throughput</em> dibuat berdasarkan data <em>traffic</em> dari API sFlow-RT dan ditampilkan menggunakan pustaka dygraph. Data topologi dan <em>traffic</em> yang tertampil pada aplikasi diperbarui setiap 10 detik. Pengujian aplikasi dilakukan dengan <em>black-box</em> <em>testing </em>menunjukkan bahwa semua fungsi pada aplikasi berhasil dilakukan. Hasil survei menunjukkan bahwa aplikasi memiliki tampilan informatif dan ramah pengguna, serta dapat memudahkan pemantauan <em>traffic</em> SDN.</p><p> </p><p><em><strong>Abstract</strong></em></p><p class="Abstract"><em>Software-defined networking (SDN) is one of the network architectures which programmable to ease network management using the controller application. SDN controllers such as ONOS only function for flow table management and do not have enough network traffic monitoring features to support network management processes. Therefore, to monitor SDN traffic other protocols such as sFlow are needed. In this research, the web-based SDN traffic monitoring system application was developed by using ONOS as SDN controller, sFlow-RT as sFlow collector, and Node.js as a web server. The results of application development produce three main features namely topology, traffic graphs, and reports. The topology visualization is based on topology data from the ONOS API and is displayed using the vis.js library. Then the throughput graph is made based on data traffic from the sFlow-RT API and displayed using the dygraph library. Topology and traffic data displayed on the application are updated every 10 seconds. Application testing is done with black-box testing showing that all functions and features of the application can function properly. The survey conducted shows that the application has an informative and user-friendly display, and can facilitate monitoring of SDN traffic.</em></p><p><em><strong><br /></strong></em></p><p> </p>

Implementation of Network Traffic Monitoring using Software Defined Networking Ryu Controller

Network traffic monitoring is vital for enhancing the overall network performance and for optimizing the traffic flows. However, an emerging growth of use in cloud services, internet-of-things, block-chain and data analytics, demand the hardware-based-network-controller to provide more features for expanding network architecture. Therefore, Software Defined Networking (SDN) offers a new solution in terms of scalability, usability and programmable software-based-network-controller for the legacy network infrastructure. In fact, SDN provides a dynamic platform for the network traffic monitoring using international standard. In this study, SDN setup and installation method uses a Mininet emulator containing a controller Ryu with switching hub component, OpenFlow switches, and nodes. The number of nodes is adding until reaches to 16 nodes and evaluated through different network scenarios (single, linear and tree topology). Findings show that the single topology gives a winning criterion compared to other topologies. SDN implementation is measured with performance parameters such as Throughput, Jitter, Bandwidth and Round-Trip Time between scenarios using the Ryu controller. Future research explores on the performance of SDN in larger network and investigates the efficiency and effectiveness of SDN implementation in mesh topology.