Muhammad Sabil Nur Raihan, . (2026) IMPLEMENTASI HORIZONTAL POD AUTOSCALER (HPA) DALAM SISTEM DETEKSI HOAKS BERBASIS ORKESTRASI KUBERNETES. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (12kB) |
|
Text
AWAL.pdf Download (909kB) |
|
Text
BAB I.pdf Restricted to Repository UPNVJ Only Download (166kB) |
|
Text
BAB II.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB III.pdf Restricted to Repository UPNVJ Only Download (892kB) |
|
Text
BAB IV.pdf Restricted to Repository UPNVJ Only Download (5MB) |
|
Text
BAB V.pdf Download (140kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (272kB) |
|
Text
DAFTAR RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (220kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (218kB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (37MB) |
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/archive.png) |
Archive
APLIKASI_2210511140_DETEKSIHOAKS.zip Restricted to Repository staff only Download (973B) |
|
Text
Jurnal_2210511140.pdf Restricted to Repository staff only Download (419kB) |
|
Text
2210511140.pdf Restricted to Repository staff only Download (7MB) |
Abstract
This study discusses the implementation of a Horizontal Pod Autoscaler (HPA) in a Kubernetes-orchestrated fake news detection system that integrates image preprocessing, Optical Character Recognition (OCR), and text classification using IndoBERT. The system was designed using a service-based architecture with asynchronous communication to reduce bottlenecks in the monolithic architecture. The implementation was deployed in local and server environments and evaluated using Locust through baseline, ramp-up, stress, and soak scenarios. The evaluation parameters included throughput, P95 latency, CPU usage, memory usage, error rate, and the number of replicas. The results show that the service-based architecture achieved higher throughput and lower P95 latency than the monolithic architecture. In the stress scenario, the local and server service configurations reached 147 req/s and 177 req/s, while the local and server monolithic configurations reached only 2.67 req/s and 0.689 req/s. The HPA increased the replicas to 25 in the local service configuration and 20 in the server service configuration. The Wilcoxon Signed-Rank Test produced a p-value of 0.0078 for throughput and P95 latency, indicating a significant performance difference. Therefore, the Kubernetes-based service architecture improved the capacity, responsiveness, and scalability of the fake news detection system.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2210511140] [Pembimbing 1: Noor Falih] [Pembimbing 2: Muhammad Panji Muslim] [Penguji 1: Indra Permana Solihin] [Penguji 2: Kharisma Wiati Gusti] |
| Uncontrolled Keywords: | Keywords: Horizontal Pod Autoscaler, Kubernetes, fake news detection, OCR, IndoBERT, performance testing. |
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | MUHAMMAD SABIL NUR RAIHAN |
| Date Deposited: | 28 Jul 2026 03:49 |
| Last Modified: | 28 Jul 2026 03:49 |
| URI: | http://repository.upnvj.ac.id/id/eprint/51724 |
Actions (login required)
 |
View Item |