Overview

Background Accommodation Services (AS) business requires Test and Production environments which have a resilient web applications tier.  AS  web applications are vulnerable to failure as they relies on a single physcial IIS web server.  

Scope Identify and evaluate available solutions that will increase the resiliency of the web tier.  The solution that best fits the business need is to be introduced to the test infrastructure in order for the solution to be tested for performance and availability in an environment reflective of Production.

Objective Provide a solution that upon single site or server hardware failure reduces the time to recover the web-tier.  In addition the revised infrastructure must be proven through load testing to not degrade web-service response times  

Phase 1 Deliverables

1.   Identify available solutions

2.   Evaluate available solutions

3.   Plan performance and availability testing (measurements and metrics)

4.   Execute perfomance testing against current Test infrastructure

5.   Introduce chosen solution to Test infrastructure and re-execute test plan

6.   Compare performance of new Test environment against 'old'

7.   Decide whether solution meets business need or whether to adopt alternative solution

8.   Introduce solution to Production environment

Phase 2 Deliverables

1.   Configure load balancer for Test

2.    Configure load balancer for Production

Benefits Aligning with the University goal to provide quality services, Accommodation Services strongly customer-focussed business requires the development of the underlying technical architecture which supports its business. There exists an operational priority to provide to Accommodation Services business with a highly resilient, available and robust web application tier. Success Criteria

·         Reduction in time to return web-services after server hardware component failure (currently 2 days)

·        Reduction in time to return web-services after network or power outage at server location (currently for duration of supply outage) 

·         Documented load testing with performance analysis showing server capabilities under increasing load and point of exponential performance deterioration.  With this information limits can be set at which vertical scaling of the web tier can be actioned

·         Test environment that mirrors production