Sciweavers

Share
CLUSTER
2007
IEEE

Non-collective parallel I/O for global address space programming models

9 years 17 days ago
Non-collective parallel I/O for global address space programming models
— Achieving high performance for out-of-core applications typically involves explicit management of the movement of data between the disk and the physical memory. We are developing a programming environment in which the different levels of the memory hierarchy are handled efficiently in a unified transparent framework. In this paper, we present our experiences with implementing efficient non-collective I/O (GPC-IO) as part of this framework. As a generalization of the Remote Procedure Call (RPC) that was used as a foundation for the Sun NFS system, we developed a global procedure call (GPC) to invoke procedures on a remote node to handle noncollective I/O. We consider alternative approaches that can be employed in implementing this functionality. The approaches are evaluated using a representative computation from quantum chemistry. The results demonstrate that GPC-IO achieves better absolute execution times, strongscaling, and weak-scaling than the alternatives considered.
Sriram Krishnamoorthy, Juan Piernas, Vinod Tippara
Added 02 Jun 2010
Updated 02 Jun 2010
Type Conference
Year 2007
Where CLUSTER
Authors Sriram Krishnamoorthy, Juan Piernas, Vinod Tipparaju, Jarek Nieplocha, P. Sadayappan
Comments (0)
books