Abbrevation
DLMCS
City
Granada
Country
Spain
Deadline Paper
Start Date
End Date
Abstract

The cost of moving data is becoming a dominant factor for performance and energy efficiency in high performance computing systems&#046; To minimize data movement, applications have to consider initial data placement and optimize both vertical data movement in the memory hierarchy and horizontal data transfer between processing units&#046;<br>While trends in computer architecture suggest that the number of computing cores on a node is continuing to increase, it is likely that some long&#8211;held programmability assumptions such as cache coherence across a whole compute node will no longer be valid on future systems&#046; At the same time, the inclusion of high&#8211;bandwidth memory and non&#8211;volatile storage will further complicate the programming of HPC systems&#046; To address this situation, application developers need to be equipped with new techniques, tools, libraries, and programming abstractions to deal with data locality as a first class concern&#046;<br>Topics of the DLMCS workshop include, but are not limited to:<br>Programming abstractions for data locality<br>Approaches for multi&#8211;level locality<br>Support for data locality in task&#8211;based programming models<br>Global address space approaches and data locality<br>Language extensions and domain&#8211;specific libraries for locality<br>On&#8211;chip networks and data locality<br>Hardware mechanisms for exploiting locality<br>Locality in large&#8211;scale HPC interconnect networks (inter&#8211;node locality)<br>Advances in cache coherence protocols and modern shared memory systems<br>Data locality and communication avoidance<br>Data locality and multi&#8211;tier memory systems<br>Approaches for processing in memory<br>Dataflow approaches<br>