Towards monitoring security policies in grid computing: A survey

Grid computing systems are complex and dynamic systems and therefore require appropriate automated management, which would enable stable and reliable operation of the whole grid environment. The research community has addressed this requirement with a number of monitoring frameworks, which serve to collect data at various levels to support decision taking and management activities within grids. However, these existing solutions seem to implement little support for collecting security-related data and enforcing appropriate security policies and constraints in this respect. With an increasing role of network connections and users remotely accessing computational resources from various locations, grid systems are no longer seen as localised and isolated ecosystems, but are coming to be more open and distributed. In this light, it is becoming more and more important to enable monitoring frameworks with capabilities to collect security-related data and check whether these observations comply with certain security constraints. Accordingly, this paper presents a survey of existing grid monitoring systems with a goal to identify an existing gap of insufficient support for handling the security dimension in grids. The survey suggests that available grid monitoring frameworks are incapable of collecting security-related data metrics and evaluating them against a set of security policies. As a first step towards addressing this issue, the paper outlines several groups of security policies, which the authors expect to be further incorporated in their own research work, and by the wider community.