IMPLEMENTATION OF MICROSERVICES ARCHITECTURE USING EVENT-DRIVEN COMMUNICATION IN EVENT TICKETING APPLICATION
Booking tickets is a dynamic needs, hence the application must be able to serve requests quickly without any issues. One of the challenges is the nature of event ticketing applications which often receive an excessive load of requests, resulting the entire application prone to downtime. In order...
Saved in:
Main Author: | |
---|---|
Format: | Final Project |
Language: | Indonesia |
Online Access: | https://digilib.itb.ac.id/gdl/view/78290 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Institut Teknologi Bandung |
Language: | Indonesia |
Summary: | Booking tickets is a dynamic needs, hence the application must be able to serve
requests quickly without any issues. One of the challenges is the nature of event
ticketing applications which often receive an excessive load of requests, resulting
the entire application prone to downtime. In order to maintain application reliability
and prevent a single point of failure, the proposed solution is to build a fault-tolerant
system by implementing a loosely-coupled microservices architecture using event-
driven communication.
Based on the functional requirements analysis of the event ticketing application, the
server can be divided into multiple microservices. However, there are ticket
booking scenarios that have dependencies between microservices to fulfill requests.
Therefore, in this final project, the communication between microservices that
supports the loosely-coupling concept will be implemented, using an event-driven
architecture with a publish-subscribe model, and data replication between
microservices.
The evaluation will be conducted by testing each ticket booking scenario with
dependency issues. The goal of the testing is to assess the loosely-coupling of each
microservice in fulfilling requests. After testing, each ticket booking scenario will
no longer have dependencies between microservices in fulfilling requests. Thus, the
loosely-coupling behaviour of the microservices in this ticketing application can be
achieved, and when one of the microservices is down, users can still perform other
actions that is handled by other active microservice (no single point of failure). |
---|