(CMU-PDL-09-111).PDFVIP

Carnegie Mellon University Research Showcase @ CMU Parallel Data Laboratory Research Centers and Institutes 11-2009 ...And Eat It Too: High Read Performance in Write- Optimized HPC I/O Middleware File Formats (CMU-PDL-09-111) Milo Polte Carnegie Mellon University Jay Lofstead Georgia Institute of Technology - Main Campus John Bent Los Alamos National Laboratory Garth Gibson Carnegie Mellon University, garth@ Scott A. Klasky Oak Ridge National Laboratory See next page for additional authors Follow this and additional works at: /pdl This Conference Proceeding is brought to you for free and open access by the Research Centers and Institutes at Research Showcase @ CMU. It has been accepted for inclusion in Parallel Data Laboratory by an authorized administrator of Research Showcase @ CMU. For more information, please contact research-showcase@. Authors Milo Polte, Jay Lofstead, John Bent, Garth Gibson, Scott A. Klasky, Qing Liu, Manish Parashar, Norbert Podhorszki, Karsten Schwan, Meghan Wingate, and Matthew Wolf This conference proceeding is available at Research Showcase @ CMU: /pdl/71 ...And eat it too: High read performance in write-optimized HPC I/O middleware ?le formats Milo Polte?, Jay Lofstead?, John Bent?, Garth Gibson?, Scott A. Klasky§, Qing Liu§, Manish Parashar?, Norbert Podhorszki§, Karsten Schwan?, Meghan Wingate?, Matthew Wolf? ?Carnegie Mellon University ?Georgia Institute of Technology ?Los Alamos National Lab §Oak Ridge National Lab ?Rutgers University 1. INTRODUCTION the Parallel Log-structured Filesystem (PLFS) [5], a stack- As HPC applications run on increasingly high process able FUSE [1] ?lesystem approach developed at Los Alamos counts on larger and larger machines, both