Designing a Cloud Architecture for SLA Requirements in Bulk Payment Case Queries: Comparative analysis

This article discusses the impact of the fourth industrial revolution on the banking sector in Indonesia, with a focus on the rise of digital banking and the introduction of the Bank Indonesia - Fast Payment (BI-FAST) system. The article proposes an architecture design that implements and compares the performance of Service-Oriented Architecture (SOA) and Event Driven Architecture (EDA) in handling bulk payment transactions in BI-FAST. The study found that both EDA and SOA can handle query information cases within the defined SLA requirement time. However, in query transaction cases with larger batch transfers, EDA with MQTT usage proved to be the more efficient, stable, and faster option, overcoming the limitations faced by SOA. EDA demonstrated faster and more stable results compared to SOA in processing query transaction cases with larger batch transfers. This efficiency can be attributed to the usage of MQTT in the EDA implementation. Therefore, for scenarios involving large numbers of batch transfers, EDA with MQTT emerges as the preferred choice for handling query transactions.