Tutorial Workshop: Big Data Analytics Without Fast Data Compromises

The Emu Memory Server is a new, scalable, shared-memory Big Data analytics system based on a fine grained, massively parallel processing architecture that employs Migratory Memory-side Processing.  In this programming model, thousands of threads execute on many cores that are closely coupled to memory.  During execution, references to non-local memory cause the hardware to automatically “migrate” the thread to the location holding the referenced data versus moving large cache blocks of data through the system. The result is greater scale, greater efficiency and lower energy required to deliver results in less time. The architecture also supports a set of remote atomic updates to non-local memory that send only the data and operation to be performed, but do not migrate the thread. This reduces network traffic from round-trip to one-way transactions, effectively avoiding bottlenecks in the network/interconnect that can cause processors to stall and applications to scale poorly.

The morning session will cover the Emu system architecture and Migratory Memory-side Processing execution model, along with an overview of the Emu programming environment, the Cilk language, and Emu-specific functionality such as data allocation and distribution, atomic operations, and thread management.  The afternoon session will be a Cilk “code, compile, and run” exercise, based on a sample application kernel. Join us and see how accessible programming for this novel architecture is. If you are familiar with C or C++, you'll quickly see how straight-forward it is to take advantage of Emu's migratory thread technology.  Please bring a laptop in order to participate in the hands-on portion of the workshop. Note: Laptops need to be able to support X11 for using the visualization tool.

For more information on Emu Technology, visit http://www.emutechnology.com/