An Investigation into the Mutability of Information in Production Grid Information Systems

Grid Information Systems are mission-critical components in today’s production Grid infrastructures. They enable users, applications and services to discover which services exist in the infrastructure and further information about their service structure and state. As the Grid Information System is pervasive throughout a Grid infrastructure, it is especially sensitive to the size of the infrastructure, and the usage of the information system increases as the infrastructure grows. Grid interoperation activities and the paradigm of multi-core processors are two main factors driving the growth of Grid infrastructures. Additional sites will bring extra resources and hence inject more information into the information system. An increasing number of cores will add to the existing execution environments available in the infrastructure, which in turn will cause an additional query load from the increased number of simultaneous computing activities that the infrastructure can support. To ensure that the current information systems are able to handle this increase in scale, it is necessary to understand where performance optimisations can be made. This paper investigates the mutability of information in the production information system of the Enabling Grid for E-Science (EGEE) project for the purpose of understanding how to best optimise the existing systems.