Western University Computer ScienceWestern Science

MSc Thesis Defense

 

Naresh Eeda

Rendering real-time dashboards using a GraphQL-based UI Architecture

 

Date:
Time:
Place:
Supervisor:
Thesis Examiners:

Extra-Departmental
Examiner:
Chair:
Wednesday, December 06, 2017
10:30 a.m.
Middlesex College, Room 320
Dr. Nazim Madhavji
Dr. Mike Bauer
Dr. Dan Lizotte

Dr. Katarina Grolinger ()
Dr. Kostas Kontogiannis

 

Abstract:

With the increase in the complexity of the systems being built and demand in the quality of service by the customers, developing and providing highly efficient real-time systems is one of the biggest challenges today for software enterprises. BluemixTM ─ IBM’s cloud offering implemented on Cloud Foundry, an open source “Platform as a Service” (PaaS), is an example of such a system. Currently, there are approx. 26 infrastructural services running in the background from where the data is fetched and is rendered on different dashboards of the system. However, the system suffers from performance issues.

This thesis explores the performance improvements of the real-time dashboards by introducing our proposed GraphQL-based UI architecture which allows caching and asynchronous loading. The test results of this architecture’s implementation on the Bluemix Usage Dashboard show that the Real data renders 245% faster and the Switching Account 153% faster than the existing system.