As HPC architecture becomes more complex, the lines between what operating and runtime systems experts call memory and the emerging off-system storage hierarchy that includes solid state devices blur. These changes result is increased complexity for application developers and increased difficulty in managing the entire process for input and output. A combination of rapid change in memory and storage technology and meeting the related requirements for the range of application classes using high performance computing (HPC) must drive the prioritization of essential new research activities in the SSIO area.
The goal of this two day workshop is to identify key challenges and define research directions that will advance the field over the next 5–7 years.
Draft - Pre Workshop Report (PDF, 1.74 MB)
ASCR Program Manager: Lucy Nowell
Rob Ross, Argonne National Lab, Co-chair
Lee Ward, Sandia National Lab, Co-chair
Glenn Lockwood, Lawrence Berkeley National Lab
Brad Settlemyer, Los Alamos National Lab
Scott Klasky, Oak Ridge National Lab
Kathryn Mohror, Lawrence Livermore National Lab
Gary Grider, Los Alamos National Lab, Senior Advisor