Abstract: A system and method for visually depicting component states in a computing system. A graphical analysis tool includes a graphical user interface for use in viewing the history of a selected component. A user selects a system component for viewing and then using a timeline “drags” the state of the selected component to various points on the timeline in order to view states of the component at the selected point in time. States of the selected component may also be played forward or backward in a continuous manner. Logging agents in the system are configured to log data corresponding to detected events and forward the log data to a central repository. The analysis tool may access the data in the repository for use in creating visual displays. Point-in-time snapshots of a component may also be stored in the central repository and used to reproduce the history.

Abstract: A method and apparatus for storing, in a fault tolerant manner, checkpoint data in backup servers. To facilitate fault tolerance, the checkpoint data produced by a production server is distributed to a plurality of backup servers. In one embodiment, the method subsegments each segment of checkpoint data, produces parity data for parity groups of the subsegmented checkpoint data, and stripes the subsegments and the parity data across a number of backup servers within a server cluster.