Analisis komparatif performa data gateway berbasis integrasi engine monitoring terpadu guna akselerasi service level agreement (sla)
Penerbit : FTI - Usakti
Kota Terbit : Jakarta
Tahun Terbit : 2026
Pembimbing 1 : Yuli Kurnia Ningsih
Pembimbing 2 : R Deiny Mardian W
Kata Kunci : Data Gateway, Engine Monitoring Integration, Infrastructure Performance, Mean Time to Repair, Servic
Status Posting : Published
Status : Lengkap
| No. | Nama File | Hal. | Link |
|---|---|---|---|
| 1. | 2026_TS_MTE_162012410001_Halaman-Judul.pdf | 14 | |
| 2. | 2026_TS_MTE_162012410001_Surat-Pernyataan-Revisi-Terakhir.pdf | 1 | |
| 3. | 2026_TS_MTE_162012410001_Surat-Hasil-Similaritas.pdf | 1 | |
| 4. | 2026_TS_MTE_162012410001_Halaman-Pernyataan-Persetujuan-Publikasi-Tugas-Akhir-untuk-Kepentingan-Akademis.pdf | 1 | |
| 5. | 2026_TS_MTE_162012410001_Lembar-Pengesahan.pdf | 1 | |
| 6. | 2026_TS_MTE_162012410001_Pernyataan-Orisinalitas.pdf | 1 | |
| 7. | 2026_TS_MTE_162012410001_Formulir-Persetujuan-Publikasi-Karya-Ilmiah.pdf | 1 | |
| 8. | 2026_TS_MTE_162012410001_Bab-1.pdf | 6 | |
| 9. | 2026_TS_MTE_162012410001_Bab-2.pdf | 17 |
|
| 10. | 2026_TS_MTE_162012410001_Bab-3.pdf | 14 |
|
| 11. | 2026_TS_MTE_162012410001_Bab-4.pdf | 55 |
|
| 12. | 2026_TS_MTE_162012410001_Bab-5.pdf | 3 | |
| 13. | 2026_TS_MTE_162012410001_Daftar-Pustaka.pdf | 5 | |
| 14. | 2026_TS_MTE_162012410001_Lampiran.pdf | 4 |
|
I Infrastruktur data gateway transaksi energi merupakan single entry point kritisbagi seluruh ekosistem transaksi pembayaran listrik dari 28 mitra perbankannasional. namun, kondisi eksisting mencatat tingkat ketersediaan (availability)aktual baru mencapai 99,71%, di bawah target service level agreement (sla)sebesar 99,9%, dengan rerata mean time to repair (mttr) mencapai 47,3 menitper insiden. analisis akar masalah menunjukkan bahwa 68,4% durasi mttr habispada fase identifikasi akibat kondisi siloed monitoring pada tim operasional(network support, infrastructure, dan application) yang beroperasi tanpamekanisme korelasi data otomatis lintas lapisan. penelitian ini menganalisisefektivitas sistem monitoring terintegrasi berbasis platform grafana yangdikembangkan oleh tim devops dengan menggabungkan sumber data heterogendari huawei imaster nce, zabbix, dan elastic stack (elk). arsitektur inimenyatukan visibilitas lintas lapisan (network layer, server health, dan logapplication) ke dalam satu dashboard tunggal menggunakan teknik timestampalignment dan normalisasi data secara real-time. evaluasi juga dilakukan terhadapmekanisme unified alerting via api telegram bot yang dikonfigurasi melaluiiterasi ambang batas (threshold) untuk mereduksi alert fatigue. hasil analisiskomparatif membuktikan bahwa korelasi lintas lapisan berhasil menurunkanmttr insiden network secara drastis sebesar 79,2% (dari 61,4 menit menjadi 12,8menit) dan meningkatkan availability menjadi 99,94%, melampaui target sla.validasi empiris menunjukkan hubungan kuat antara anomali fisik network dankegagalan application dengan koefisien korelasi pearson r = -0,89 (p < 0,001)antara parameter degradasi daya optik (optical power degradation) dan lonjakanhttp 5xx error rate. penelitian ini memberikan kontribusi berupa validasi empirismodel analisis korelasi lintas lapisan dalam menurunkan nilai mttr secarakonsisten pada infrastruktur transaksi energi nasional.
T The energy transaction data gateway infrastructure serves as a critical singleentry point for the entire electricity payment transaction ecosystem across 28national partner banks. however, the existing condition recorded an actualavailability rate of only 99.71%, which fell below the service level agreement(sla) target of 99.9%, with an average mean time to repair (mttr) of 47.3minutes per incident. root cause analysis indicated that 68.4% of the mttrduration was consumed during the identification phase due to siloed monitoringwithin operational teams (network support, infrastructure, and application)operating without cross-layer automated data correlation mechanisms. this studyanalyzes the effectiveness of an integrated monitoring system based on the grafanaplatform developed by the devops team by combining heterogeneous data sourcesfrom huawei imaster nce, zabbix, and elastic stack (elk). this architectureunifies cross-layer visibility (network layer, server health, and application logs)into a single dashboard utilizing real-time timestamp alignment and datanormalization techniques. an evaluation was also conducted on the unified alertingmechanism via the telegram bot api, configured through threshold iterations tomitigate alert fatigue. comparative analysis results demonstrate that cross-layercorrelation successfully slashed the network incident mttr by 79.2% (from 61.4minutes down to 12.8 minutes) and boosted availability to 99.94%, therebyexceeding the sla target. empirical validation reveals a strong relationshipbetween network physical anomalies and application failures with a pearsoncorrelation coefficient of r = -0.89 (p < 0.001) between the optical powerdegradation parameter and spikes in the http 5xx error rate. this researchcontributes an empirical validation of a cross-layer correlation analysis model inconsistently reducing mttr values across the national energy transactioninfrastructure.