Claim Missing Document
Check
Articles

Load Balancing Server Web Berdasarkan Jumlah Koneksi Klien Pada Docker Swarm Dimas Setiawan Afis; Mahendra Data; Widhi Yahya
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 1 (2019): Januari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (333.817 KB)

Abstract

Most web servers that are used today still use single backend server architectures. The problem that arises is how the single server is able to handle large data requests. We must consider using web server clustering to improve web server reliability. We built this cluster using virtualization technology such as virtual containers. One of the container-based virtualisations currently is Docker. However, managing multiple containers to make a single service is a challenging task. Docker introduces a distributed system development tool called Docker Swarm. We propose a load balancing mechanism on Docker Swarm to balance internal load, so that it can distribute requests to web server. In addition, we also try to find out the performance of roundrobin and leastconn algorithms on loadbalancing mechanisms that will be used on Docker Swarm. The test results show that loadbalancing applying the least connection algorithm has a throughput of 15 Mbps in 1000 requests, 17 Mbps in 3000 requests, 17 Mbps in 5000 requests, while the round robin algorithm has a throughput of 15 Mbps in 1000 requests, 14 Mbps in 3000 requests, 15 Mbps in 5000 requests. The results show the least connection algorithm has better performance than the round robin algorithm. In addition, the results of data distribution are balanced on each available web server.
Implementasi Switch Openflow Berbasis Software Dengan Memanfaatkan Raspberry Pi Untuk Infrastruktur SDN Imam Santoso; Mochammad Hannats Hanafi Ichsan; Widhi Yahya
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 1 (2019): Januari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (713.699 KB)

Abstract

Software-Defined Networking (SDN) is a dynamic, manageable, adaptable and cost-efficient architecture. The Raspberry software-based openflow switch is currently untested, so it still can not be the right reference to replace dedicated openflow switch devices. This study aims to implement OpenvSwitch to be applied to Raspberries so as to make it an openflow switch that can be used within the SDN network and then test the performance of Raspi switches, throughput, packet loss and delay. The results obtained from this research that Raspberry openflow switches can be used to replace dedicated openflow switches on a small scale such as office scope but not for large scale companies like companies due to bandwidth limitation on USB to Ethernet adapters, as well as RAM from Raspberry. Based on the results of throughput testing, throughput will decrease as the number of clients increases. While based on packet loss test results, packet loss will increase significantly when the number of clients more than 3. Then based on delay testing, the first test has a longer delay compared to the next test, this is because in the first ICMP (ping) packet, the ARP process occurs while in the second and so on does not do the ARP process because the information about ARP already exists in the ARP cache.
Pengembangan Stateful Packet Inspection dengan Metode NDPMon untuk Duplicate Address Detection Steven Urbani; Widhi Yahya; Eko Sakti Pramukantoro
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 2 (2019): Februari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1345.455 KB)

Abstract

Software defined networking is a new field of research, where the concept aims to replace an inflexible and complex network with an innovative and agile network by separating the control plane and data plane. One of the protocols that can carry the concept of SDN is OpenFlow. Problems about network security that related about gap on conventional network are also occur on the OpenFlow network. One of the network attacks which can be threatening the security of the OpenFlow network is DoS of duplicate address detection, which attackers exploit a gap on IPv6 duplication checking schemes in a local network. Program that can be applied on controller to resolve some security issues in OpenFlow is Stateful Packet Inspection. To resolving the gap, development of the Stateful Packet Inspection can be done to prevent attacks on duplicate address detection processes by adding an NDP checking mechanism using the NDPMon method. There are several tests to test the system's functionality and performance in overcoming gaps in DAD process on the OpenFlow network. From the test results, we obtained that the system can noticed duplicate address detection process and successed to handle three test scenarios for DAD DoS attacks.
Implementasi Penyimpanan Data Persisten pada Docker Swarm Menggunakan Network File System (NFS) Andreas Frederius; Mahendra Data; Widhi Yahya
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 2 (2019): Februari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (409.144 KB)

Abstract

Docker Swarm is a distributed system development technology for managing the Docker engine group. With Docker Swarm you can run multiple containers at the same time in the Docker engine group. In a distributed method using Docker Swarm a persistent data storage is needed. But facing Docker Swarm stores data on the container, if the container is deleted then the data will also be deleted. Therefore we need a persistent alternative storage data. Network File System (NFS) is an open protocol that can be used to share files on many computer networks and operating systems. The design of the NFS architecture on Docker Swarm uses a client-server architecture. Docker Swarm as a client and NFS as a server. NFS is able to provide persistent storage data on Docker Swarm by configuring data even if the container and machine are restarted. NFS can prepare data on Docker Swarm to retrieve data that has been stored on NFS-3 data remains persistent. The average write speed performance on NFS is 30,168 KB while the working speed of reading the average on NFS is 63,939 KB.
Analisis Kinerja OpenFlow Switch Aruba 2920 Zefanya Hermanta; Widhi Yahya; Achmad Basuki
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 2 (2019): Februari 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (396.198 KB)

Abstract

OpenFlow switches are network switches based on the OpenFlow protocol in forwarding packets on an SDN network. The various switch vendors have provided OpenFlow-enabled networks to support various versions of the OpenFlow protocol. In the implementation of switch there are capacity limitations of TCAM on line-speed forwarding and only accommodate a limited number of streams. Therefore, it will produce different performance. In this study, the authors conducted an analysis of performance OpenFlow switch Aruba 2920. In addition, it uses the emulator mininet as a reference for comparing network performance. The test results show that the system can search several paths in the network topology and along with the path formation time or referred to as convergance time. Tests show that throughput and packet loss on multi-path routing produce better performance, but convergence time is slightly longer than single-path routing. The resulting performance on the Aruba 2920 OpenFlow switch network is very low compared to the emulator mininet, because there are limitations to the group table type select and thrash packet headers on the hardware table so that the flow table is managed in software table.
Perbandingan Routing Ulang Pada Algoritme Dijkstra dan Floyd-Warshall Dalam Mengatasi Link Failure Pada Arsitektur SDN Risailin Dwi Jaka Fauzi; Rakhmadhany Primananda; Widhi Yahya
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 3 (2019): Maret 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (948.673 KB)

Abstract

There are 2 (two) types of routing, static routing and dynamic routing. Static routing is a type of routing that will done manually, while dynamic routing is a type of routing that is more automatic routing. One of network architecture that implements dynamic routing at this time is SDN, in the routing process there are several problems, one of that problem is link failure, where failure on a path can affect data sending process in the network. At this time there are many types of routing algorithms such as Dijkstra's Algorithm, Dijkstra's algorithm way to finding the shortest path by checking each path until it produces the best path. While Floyd-Warshall is more dynamic algorithm in searching for the shortest path not just fixed on the lowest cost value, but also considering the future consequences of each step that will be taken. To know about performance comparison the researcher made a comparison on the re-routing process using the Djikstra and Floyd-Warshall algorithms to find out which ones were faster in overcoming the link failure at SDN. From the comparison of the Djikstra and Floyd-Warshall algorithm's convergence time, Dijkstra's algorithm has better results i.e for scenario 1 0ms, scenario 2 16.401ms, and scenario 3 17.200ms compared to Floyd-Warshall algorithm, for scenario 1 0ms, scenario 2 19.803 ms, and scenario 3 20.401ms.
Implementasi Load Balancing Server Web Berbasis Docker Swarm Berdasarkan Penggunaan Sumber Daya Memory Host Mohamad Rexa Mei Bella; Mahendra Data; Widhi Yahya
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 4 (2019): April 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1558.783 KB)

Abstract

Virtualization based container is very popular among programming development because this virtualization is lightweight virtualization where Linux kernel allow to share resource between containers so performace container does not interfere with each other. One of the most used virtualization based container is Docker. Docker is open source software which can be change as we wish. Docker container can be used to clustering web server. This method can be used to decrease a single point of failure (SPOF) in web server. However to manage a lot container is complex, but Docker have engine to solve this problem called Docker Swarm. Docker Swarm have internal load balancing but just to manage between container and can't be monitored.So can make resource between host is not unequal distributed. Therefore this research purpose is to distribute web server traffic inside a Docker swarm using loadbalancing based on the resource utilization of the host machine. There are several tests to test the system's functionality and performance in overcoming gaps in failover based time and load balancer based memory. From the test results, we obtained that failover based time and loadbalancer based memory can work in Docker Swarm and can solved the problem about unequal distributed between host.
Analisis Kinerja Dan Konsumsi Sumber Daya Aplikasi Web Server Pada Platform Raspberry Pi Andhika Dwitama Putra; Widhi Yahya; Adhitya Bhawiyuga
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 4 (2019): April 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (618.108 KB)

Abstract

Nowadays, the utilization of single board computer in the current digitalitation era is developing rapidly. One example of Single Board Computer is Raspberry Pi. The limited hardware specification makes the resource allocation consumed by Raspberry Pi also limited. Web server application has a role in affecting resources performance and consumption in the Raspberry Pi platform. This research aims to figure out which web server that has a good resources performance and consumption in Raspberry Pi. The test method done includes the resources performance and consumption test by using several parameters: throughput, average response time, cpu and memory power usage, while the web server application compared including Apache, Nginx, Lighttpd, Apache Tomcat, and Jetty. The test is using user as an input parameter in the amount of 250, 500, 1000 and 1200. Based on the test result as well as the analysis, Nginx have the best performance based on the highest score of throughput with 357 KB/s and the lowest score of average response time with 1117 ms in serving 1200 user and using data to the test in the amount of 1.536,1 Kbytes and Lighttpd have the best resources consumption based on the lowest score of cpu usage with 4,46 % and the lowest score of memory usage with 22,16 % in serving 1200 user and using data to the test in the amount of 1.019,1 Kbytes.
Implementasi Privacy Module Menggunakan Algoritme Enkripsi Trivium Pada SNMPL8 Dimas Dwi Radhityo; Ari Kusyanti; Widhi Yahya
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 4 (2019): April 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (750.517 KB)

Abstract

Device management is needed when more internet of things (IoT) devices are connected in a network. One of the device management protocols is SNMP. But SNMP generates large data. Therefore there is a monitoring system that developed for effectivity in generating data called SNMPL8. On the other hand, SNMPL8 did not focus on security functions. One of the security functions for monitoring systems is the SNMPv3's Privacy Module. However, the Privacy Module still using CBC-DES algorithm that is not recommended by NIST, so a replacement algorithm is needed. One of the candidates is Trivium algorithm that designed for constrained devices. Therefore, this study uses the Trivium algorithm adopted on the SNMPv3's Privacy Module for network monitoring systems. Based on the getResponse testing, the Privacy Module requires 17.28 ms to generate keystream and 2.76 ms for the encryption process. The difference in memory at implementation is 176 Bytes, while the difference in the total data package is 769 Bytes.
Implementasi Algoritme BLAKE2S pada Protokol MQTT Sandy Ramadhan Wijaya; Ari Kusyanti; Widhi Yahya
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 4 (2019): April 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (781.239 KB)

Abstract

Message Queue Telemetry Transport (MQTT) is a protocol with a publish-subscribe architecture for Internet of Things (IoT) that designed to be a lightweight protocol when applied. MQTT also a protocol with low bandwidth usage because of the small size transmitted data. To secure the integrity of transmitted data from alteration, substitution, and insertion, an integrity checking mechanism of the received data is required. Because the received data could affect the results of data processing like storing data to the database and decision making by the control unit. Although MQTT can be applied using Secure Sockets Layer (SSL), but SSL requires a lot of resources. As an alternative to the SSL's implementation, Message Authentication Code (MAC) is used in this study for an integrity checking mechanism. There is a MAC algorithm that has been widely used, such as the HMAC-MD5 and HMAC-SHA1 algorithms. However, the MD5 and SHA1 algorithms are insecure because they are vulnerable to second preimage and collision attacks. So as an alternative BLAKE2S algorithm is used because it is resistant to second preimage and collision attack attacks. Based on the test result of the implemented BLAKE2S algorithm to systems with the MQTT protocol, the system knows if there is an alteration, substitution, and insertion of the received data with an average of memory usage overhead only 0.0041 MB in the publisher side and 0.0044 MB in the subscriber side.
Co-Authors Achmad Basuki Adhitya Bhawiyuga Adi Iman Utama Aditya Bhawiyuga Afredy Carlo Sembiring Agung Nugroho Ahmad Naufal Romiz Ahmad Rizal Muttaqin Andhika Dwitama Putra Andika Kurniawan Andreas Frederius Aprilia Kartika Sriastunti Aprillia Arum Pratiwi Ardy Frayogi ari kusyanti Arief Budi Setiawan Arnandha Hadi Pranata Aswin Suharsono Bagas Prawira Adji Wisesa Bagus Prasetyo Bagus Satria Wiguna Bhaskara Vito Tarigan Dany Primanita Kartikasari Dany Rahmana Dea Asmara Gita Delarta Tok Adin Dimas Dwi Radhityo Dimas Setiawan Afis Doni Hadiyansyah Enggar Saka Dirgantara Faizal Ramadhan Fariz Andri Bakhtiar Florensius Tri Putra Simamora Frondy Fernanda Ferdianto Gabreil Arganata Galeh Prehandayana Guntur Wahyu Pamungkas Hafidz Rizal Hafizhul Karim Hasbi Razzak Heru Nurwarsito Hilkya Dwima Palembangan Ibrahim Attamimi Imam Santoso Ismiarta Aknuranda Kasyful Amron Lalu Fani Islahul Ardy Landika Hari Suganda Lucky Ronny Chandra Negara M. Ali Fauzi Mahendra Data Maskur Purwiadi Moch. Wahyu Imam Santosa Mochammad Hannats Hanafi Mochammad Hannats Hanafi Ichsan Moh Wahyudi Putra Mohamad Rexa Mei Bella Mohammad Hafidzar Rakhman Muhammad Aji Wibowo Muhammad Luthfi Muhammad Rifqi Muzaki Muhammad Sholeh Muharrom Abdillah Niki Yuniar Wicaksono Nur Fauzi Panser Brigade Muhammad Poltak G. Hutajulu Pramukantoro, Eko Sakti Primantara Hari Trisnawan Rahadiyan Yuniar Rahmanda Rakhmadhany Primananda Ratih Kartika Dewi Reza Andria Siregar Rian Bayu Pambudi Risailin Dwi Jaka Fauzi Riski Julianto Romy Dwi Andika Manullang Sabriansyah Rizqika Akbar Sandy Ramadhan Wijaya Steven Urbani Tista Pamungkas Ragil Alit Uis Yudha Tri Wirawan Ulfa Khoirul Azizah Wijaya Kurniawan Yasin Rizqi Afandi Zavero Brillianata Abilovani Zefanya Hermanta