Summary of what we have covered¶

• Simple remote execution of functions, or statements
• Distribution of data
• Parallel map
• Multiplexed execution
• Load-Balanced task farming
• Asynchronous task submission and result retrieval
• More efficient data movememnt with memmap and MPI

And further things we have not covered¶

• Functional Task Dependencies
• Configuring the IPython cluster (working with profiles, SGE, MongoDB, etc.)
• Tuning the IPython Cluster (TaskScheduler.hwm, custom serialization, etc.)
• Inter-engine communication with PyZMQ or MPI
• decorators for parallel and remote functions
• Result caching, and task metadata
• The TaskResult Database
• Task resubmission
• Queue monitoring and management

See the IPython.parallel docs for more detailed coverage of these topics, and check out our examples.