Abstract: A system and method for facilitating an automatic rolling upgrade of distributed software. In a set of computing devices (e.g., a cluster), each node executes the distributed program, but different nodes may operate any number of different versions of the program. During a rolling upgrade of the software, the nodes all operate according to a specified acting version of the software. A node attempting the join the cluster may be required to have a software version at least equal to the acting version. When all nodes have been upgraded to a version newer than the current acting version, the acting version is reset to the lowest software version installed on a node. The rolling upgrade completes automatically when all nodes run at a specified software version (e.g., the most recent) or have the same version installed.

Abstract: A method and mechanism for managing data to resolve incompatibilities between different versions of data is disclosed. According to one approach, incompatibilities to private data is managed using group/member dependent key(s)/link(s), static shared data is managed using group/member dependent key(s)/link(s), and dynamic shared data is managed using data transformation function(s).

Abstract: The method termed WaveCluster for mining spatial data. WaveCluster considers spatial data as a multidimensional signals and applies wavelet transforms, a signal-processing technique, to convert the spatial data into the frequency domain. The wavelet transforms produce a transformed space where natural clusters in the data become more distinguishable. The method quantizes a feature space to determine cells of the feature space, assigns objects to the cells, applies a wavelet transform on the quantized feature space to obtain a transformed feature space, finds connected clusters in sub bands at different levels of the transformed feature space, assigns labels to the cells, creates a look-up table, and maps the objects to the clusters. The method can manage spatial data in a two-dimensional feature space. The method also is applicable to a feature space that is made up of an image taken by a satellite.

Abstract: A mechanism is disclosed for managing a distributed computing system in which a plurality of services are provided by a plurality of members. Initially, one of the members of the system is elected as the global master for the system. Thereafter, the global master selects a service master (where a service master comprises a member of the system) for each of the services provided by the system, such that each service has an associated service master. Each service master may be selected based upon an associated set of service master selection criteria. Once selected, each service master manages requests for the service with which the service master is associated. In this manner, services and the requests for the services are effectively managed.

Abstract: Disclosed is a method, mechanism, and computer usable medium for managing and accessing static and shard data. In one approach, described is a method and mechanism for implementing group dependent keys (GDKs) in a computing system, in which the GDK is visible to all members of a distributed system, but its value(s) and subtree(s) could be different for different groups. Members of each group see the same view of the value and subtree of a GDK. Also disclosed is a method, mechanism, and computer usable medium for implementing group dependent links (GDLs) in a computing system. According to one approach, a data transformation function is used to coordinate changes to different versions of shared data.

Abstract: A system and techniques are described for managing key-value pairs using an architecture that does not restrict its users to any specific platform or storage subsystem. According to one aspect of the invention, the repository is portable in that the architecture need not change based on the environment and platform in which the repository is used. Rather, the portion of the architecture that is platform-specific is limited to plug-ins at a storage abstraction layer of the repository. The plug-ins expose the same storage abstraction interface to the other layers of the repository, but are implemented to interact with different platforms and storage subsystems. Consequently, in response to being ported from one platform to another, the repository may simply change the plug-in to invoke for persistent storage operations.