Dapper, or "Distributed and Parallel Program Execution Runtime", is a tool for taming the complexities of developing for large-scale cloud and grid computing, enabling the user to create distributed computations from the essentials: the code that will execute, along with a dataflow graph description. It supports rich execution semantics, carefree deployment, a robust control protocol, modification of the dataflow graph at runtime, and an intuitive user interface.
|Tags||Distributed Computing Scientific Computing cloud computing Workflow Frameworks|
|Operating Systems||OS Independent|
Release Notes: The ServerLogic#closeIdleClients method has been changed to better match the user's intuitive notion of idleness. A user option for specifying the server's hostname has been added. Networking internals have been reworked to use new APIs. The build process has been updated to support both 32- and 64-bit Windows cross-compilation. The dapper.* hierarchy has been renamed to org.dapper.*.
Release Notes: The bipartite matching internals have been rewritten to use a different algorithm implementation. An API method exposing the number of pending computations has been introduced. The FlowListener callback interface has been removed and replaced with an event-based notification system. An Ubuntu package is now available.
Release Notes: The FlowListener abstraction has been added, and enables users to associate arbitrary metadata with dataflows and their nodes. A memory leak concerning defunct dataflows has been fixed. The build process is now fully integrated with Apache Ivy. Java 1.6 is now required to build and run. More complete documentation has been generated Doxygen and apiviz have been added.
Release Notes: A new, flexible logging infrastructure has been added. Initializers for logging structures have been moved out of the Server and Client classes and into drivers. Finite state machines have been updated to the new annotation-driven API. The source code has been normalized to have 8 spaces instead of tabs.
Release Notes: Command line options for the client and server are now available, courtesy of the Apache Commons CLI library. The client process lifecycle is now defined as ending when a disconnect from the server happens. Stem generation functionality has moved from being a member method of OutputHandleResource to being a static method of CodeletUtilities. The FlowNodeFactory class is used in favor of direct instantiation of FlowNodes. Building of native components has migrated to the CMake. Logging has migrated to SLF4J. A README has been added to all distributions.