High Availability Systems

Computer systems are often distributed across a network to provide services to the end-users. These systems must be available and provide services to the users when required. For this reason, high availability system technologies have captured the attention of IT-organizations. Most companies consider it important to provide continuous services with minimal downtime to the end-users. The implementation of service availability is a complex task with multiple constraints, including security, performance, and system scalability. The first chapter of the thesis introduces the high availability system and the objectives of the thesis. In the second, third, fourth and fifth chapters, concepts, redundancy models, clusters and containers are described. In the sixth chapter, an approach to measure the availability of the components of IT-system using the Application Availability Measurement method is provided. The seventh and eighth chapters contain a case study. The seventh chapter contains actual backup system design overview, issues related to the current methods and tools to measure the availability used by a Finnish software company. In the eighth chapter, as part of the case study, a solution design is proposed based on the principle of service delivery decomposition into a set of measurement points for service level indicators. A plan is provided to show how to implement a method and provide tools to measure the availability of the backup system used by a Finnish software company