PERANCANGAN INFRASTRUKTUR REDUNDANSI DENGAN MENERAPKAN HIGH AVAILABILITY UNTUK MENJAMIN KEANDALAN OPERASIONAL BISNIS

Abstract

The availability of information technology services is crucial for business operations, but infrastructure with a Single Point of Failure risks causing downTime that disrupts services. This research focuses on the design and implementation of High Availability-based redundancy infrastructure to maintain service reliability when a virtualization host or network is disrupted. The research method uses the Network Development Life Cycle, which includes requirements analysis, design, simulation/prototype, implementation, monitoring, and management. The implementation was carried out in a VMware vSphere environment with two ESXi hosts managed through vCenter, and iSCSI-based NetApp shared storage as a shared datastore to support the failover mechanism and automatic VM Recovery. Testing includes host failover, network failover, and Failback scenarios with connectivity monitoring. The results show that in the host failover scenario, the virtual machine was successfully recovered automatically with 33 seconds of downTime, while in the network failover and Failback scenarios, the service remained accessible with no observed downTime (0 seconds). The conclusion of this research shows that the design and implementation of High Availability can improve service reliability, minimize downTime during disruptions, and support achieving the 99.9% SLA target in the test scenario.

Ketersediaan layanan teknologi informasi sangat penting bagi operasional bisnis, namun infrastruktur dengan Single Point of Failure berisiko menyebabkan downTime yang mengganggu layanan. Penelitian ini berfokus pada perancangan dan implementasi infrastruktur redundansi berbasis High Availability untuk menjaga keandalan layanan saat terjadi gangguan pada host virtualisasi atau jaringan. Metode penelitian menggunakan Network Development Life Cycle yang meliputi analisis kebutuhan, perancangan, simulasi/prototipe, implementasi, monitoring, dan manajemen. Implementasi dilakukan pada lingkungan VMware vSphere dengan dua host ESXi yang dikelola melalui vCenter serta shared storage NetApp berbasis iSCSI sebagai datastore bersama untuk mendukung mekanisme failover dan pemulihan otomatis VM. Pengujian mencakup skenario failover host, failover jaringan, dan Failback dengan pengamatan konektivitas. Hasil penelitian menunjukkan bahwa pada skenario failover host, virtual machine berhasil dipulihkan secara otomatis dengan downTime 33 detik, sedangkan pada skenario failover jaringan dan Failback layanan tetap dapat diakses tanpa downTime teramati (0 detik). Kesimpulan penelitian ini menunjukkan bahwa rancangan dan implementasi High Availability mampu meningkatkan keandalan layanan, meminimalkan downTime saat gangguan, serta mendukung pencapaian target SLA 99,9% pada skenario pengujian