ANALISIS THROUGHPUT DAN RESPONSE TIME POLLING API MENGGUNAKAN DYNAMIC LOAD BALANCING SOFTWARE-BASED

KHOIRULLOH, HILMI (2024) ANALISIS THROUGHPUT DAN RESPONSE TIME POLLING API MENGGUNAKAN DYNAMIC LOAD BALANCING SOFTWARE-BASED. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (HAL COVER) 01 COVER.pdf Download (699kB) | Preview
	Text (ABSTRAK) 02 ABSTRAK.pdf Download (33kB)
	Text (BAB I) 03 BAB 1.pdf Restricted to Registered users only Download (93kB)
	Text (BAB II) 04 BAB 2.pdf Restricted to Registered users only Download (329kB)
	Text (BAB III) 05 BAB 3.pdf Restricted to Registered users only Download (140kB)
	Text (BAB IV) 06 BAB 4.pdf Restricted to Registered users only Download (239kB)
	Text (BAB V) 07 BAB 5.pdf Restricted to Registered users only Download (73kB)
	Text (DAFTAR PUSTAKA) 08 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (125kB)
	Text (LAMPIRAN) 09 LAMPIRAN.pdf Restricted to Registered users only Download (700kB)

Abstract

In this digital era, information is expected to be distributed quickly and accurately. Therefore, it is crucial to maintain the stability of a platform to ensure the availability and reliability of the services provided to users. However, uneven load on existing service servers can result in suboptimal server performance. This issue can be addressed by using a load balancer. The function of a load balancer is to distribute connection loads to servers to enhance server performance. One scheme of load balancing is called dynamic load balancing, an example of which is the least connection method. This method works by selecting the server with the lowest outgoing connections. Load balancers can be used in several software load balancers such as Nginx, HAProxy, and Envoy. In this study, an analysis was conducted on the impact of using dynamic load balancing software-based on the average throughput and response time of the polling API service on BigVision startup service servers. The research conducted is a mixed-methods research with several supporting research methods, namely, interviews, literature studies, and observations. The data source for this research is the load test results on 2 BigVision service servers. The testing scheme was carried out using 3 connection numbers with 2 different CPS configurations. The test results show that the average throughput and response time of the polling API service have improved in performance. Therefore, it can be concluded that using dynamic load balancing software-based can improve server performance, thus ensuring that a service or platform remains stable. Keyword : Server Performance, Load Balancer, Software Load Balancer, Dynamic Load Balancer, Polling Dalam era digital ini informasi diharapkan dapat tersebar secara cepat dan akurat. Oleh karena itu, pentingnya menjaga kestabilan sebuah platform untuk menjaga availability dan reliability terhadap layanan yang diberikan kepada pengguna layanan. Namun, load yang tidak merata terhadap server layanan yang ada, menyebabkan kurang maksimalnya performa server tersebut. Untuk mengatasi masalah tersebut dapat dilakukan dengan menggunakan load balancer. Fungsi load balancer adalah untuk mendistribusikan beban koneksi terhadap server untuk meningkatkan performa server layanan. Salah satu skema load balancer dinamakan skema dynamic load balancer dengan contoh metodenya adalah metode least connection yang memlilki cara kerja dengan memilih server dengan koneksi outgoing paling rendah. Penggunaan load balancer sendiri sudah dapat digunakan dalam beberapa software load balancer seperti Nginx, HAProxy, dan Envoy. Pada penelitian ini, dilakukan analisis mengenai pengaruh penggunaan dynamic load balancing software-based terhadap nilai rata-rata throughput dan response time terhadap service polling API di server layanan startup BigVision. Penelitian yang dilakukan adalah penelitian campuran dengan dilakukan beberapa metode penunjang penelitian yaitu, wawancara, studi literatur, dan observasi. Data yang dijadikan sumber dari penelitian ini adalah hasil load test pada 2 server layanan BigVision. Skema pengujian dilakukan dengan menggunakan 3 jumlah koneksi dengan 2 konfigurasi CPS yang berbeda. Hasil pengujian menunjukkan bahwa nilai rata-rata throughput dan response time terhadap service polling API mengalami peningkatan secara performa. Sehingga dapat disimpulkan bahwa dengan menggunakan dynamic load balancing software-based dapat meningkatkan performa dari server sehingga sebuah layanan atau platform tetap stabil. Kata kunci : Performa Server, Load Balancer, Software Load Balancing, Dynamic Load Balancing, Polling

Actions (login required)

View Item