High Availability Testing - Stories from Enterprise and Consumer Services

摘要

Quality of a cloud service is more than having a defect-free service which meets the requirements. If the product or service has downtime or is slow in recovering from faults, it will directly impact the product or service adoption and customer satisfaction. High availability is often an implied need; but customers expect it to be there and its absence will impact the business. With global users and mobile devices using cloud services, testing for high availability, also known as fault tolerance, becomes much more important. One useful measurement to gauge service availability is customer feedback and the customer perceived downtime. If the user feels the cloud service is erroneous or has frequent downtime, it does not matter what process or tools are used for high availability, the customer is always right. This document explains high availability testing with fault model techniques that can be used to improve the availability of a cloud service (enterprise or consumer) and to plan for the right thing, and get the right things for the right user at the right time - every time.