Lumentut, Aloysius Juan Farrel (2026) Pengembangan Sistem Pengujian Reliabilitas Aplikasi Microservices menggunakan LitmusChaos. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5025221176-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5025221176-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy |
Abstract
Adopsi arsitektur microservices menjadi solusi untuk memenuhi kebutuhan sistem yang skalabel dan andal, namun kompleksitas menjadi tantangan dalam pengujian reliabilitas karena metode pengujian tradisional belum mencakup seluruh kondisi pada dunia nyata. Penelitian ini menerapkan paradigma Chaos Engineering memanfaatkan LitmusChaos sebagai metode evaluasi reliabilitas aplikasi microservices melalui injeksi kegagalan terkontrol. Metode yang dirancang mencakup penyusunan kerangka evaluasi reliabilitas dengan memetakan sub-karakteristik reliabilitas ISO/IEC 25010, penetapan indikator pengukuran melalui SLI dan SLO, serta pemilihan skenario chaos yang relevan. Sistem yang dibangun terdiri atas dua komponen yakni orkestrator sebagai pengatur utama proses pengujian, dan kluster lingkungan pengujian. Pengujian dijalankan pada lima scenario yang merepresentasikan aspek reliablitas ISO/IEC 25010, yakni skenario operasi normal, availability terdegradasi, faultlessness terdegradasi, fault tolerance terdegradasi, dan recoverability terdegradasi, sebanyak lima iterasi. Hasil menunjukkan error rate naik 715% pada aspek faultlessness, system availability turun 5,836% pada aspek availability, success rate layanan pemanggil turun 67,75% pada aspek fault tolerance, dan waktu pemulihan meningkat 550% pada aspek recoverability. Hasil pengujian selaras dengan skenario degradasi yang dirancang, mengindikasikan sistem mampu mengukur dan membedakan perilaku sistem lewat indikator reliabilitas. Temuan lain mengungkap bahwa indikator error rate dan availability terdegradasi signifikan dan sensitif terhadap jenis kegagalan jaringan. Sedangkan indikator pengukuran response time sensitif terhadap jenis kegagalan tekanan sumber daya komputasi serta gangguan pod-network-latency.
====================================================================================================================================
The adoption of microservices architecture has emerged as a solution to meet the need for scalable and reliable systems. However, its complexity poses a challenge for reliability testing, as traditional testing methods do not cover all real-world conditions. This research applies the Chaos Engineering paradigm using LitmusChaos as a method for evaluating the reliability of microservices applications through the injection of controlled failures. The designed method includes the development of a reliability evaluation framework by mapping the reliability sub-characteristics of ISO/IEC 25010, the establishment of measurement indicators through SLIs and SLOs, and the selection of relevant chaos scenarios. The system consists of two components, an orchestrator, which serves as the primary controller of the testing process, and a testing environment cluster. Testing was conducted across five scenarios representing the reliability aspects defined in ISO/IEC 25010 namely, normal operation, degraded availability, degraded faultlessness, degraded fault tolerance, and degraded recoverability with five iterations each. The results showed a 715% increase in the error rate for the faultlessness aspect, a 5.836% decrease in system availability for the availability aspect, a 67.75% decrease in the caller service success rate for the fault tolerance aspect, and a 550% increase in recovery time for the recoverability aspect. The test results align with the designed degradation scenarios, indicating that the system can measure and distinguish system behavior through reliability indicators. Other findings reveal that the error rate and availability indicators degrade significantly and are sensitive to the type of network failure. Meanwhile, the response time measurement indicator is sensitive to failures caused by computational resource pressure and pod-network-latency disruptions.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Reliabilitas, Microservices, Chaos Engineering, LitmusChaos, ISO/IEC 25010, SLI, SLO, Reliability, Microservices, Chaos Engineering, LitmusChaos, ISO/IEC 25010, SLI, SLO |
| Subjects: | Q Science > QA Mathematics > QA76.754 Software architecture. Computer software Q Science > QA Mathematics > QA76.758 Software engineering |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Aloysius Juan Farrel Lumentut |
| Date Deposited: | 28 Jul 2026 01:06 |
| Last Modified: | 28 Jul 2026 01:06 |
| URI: | http://repository.its.ac.id/id/eprint/137916 |
Actions (login required)
 |
View Item |