BEGIN:VCALENDAR PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN VERSION:1.0 BEGIN:VEVENT DTSTART:20121111T153000Z DTEND:20121112T000000Z LOCATION:355-B DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: I/O on HPC systems is a black art. This tutorial sheds light on the=0Astate-of-the-art in parallel I/O and provides the knowledge necessary for=0Aattendees to best leverage I/O resources available to them. We cover the=0Aentire I/O software stack from parallel file systems at the lowest layer, to=0Aintermediate layers (such as MPI-IO), and finally high-level I/O libraries=0A(such as HDF-5). We emphasize ways to use these interfaces that result in=0Ahigh performance, and benchmarks on real systems are used throughout to show=0Areal-world results.=0A=0AThis tutorial first discusses parallel file systems in detail (PFSs). We cover=0Ageneral concepts and examine four examples: GPFS, Lustre, PanFS, and=0APVFS. We examine the upper layers of the I/O stack, covering POSIX I/O,=0AMPI-IO, Parallel netCDF, and HDF5. We discuss interface features, show code=0Aexamples, and describe how application calls translate into PFS=0Aoperations. Finally we discuss I/O best practice. SUMMARY:Parallel IO In Practice PRIORITY:3 END:VEVENT END:VCALENDAR BEGIN:VCALENDAR PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN VERSION:1.0 BEGIN:VEVENT DTSTART:20121111T153000Z DTEND:20121112T000000Z LOCATION:355-B DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: I/O on HPC systems is a black art. This tutorial sheds light on the=0Astate-of-the-art in parallel I/O and provides the knowledge necessary for=0Aattendees to best leverage I/O resources available to them. We cover the=0Aentire I/O software stack from parallel file systems at the lowest layer, to=0Aintermediate layers (such as MPI-IO), and finally high-level I/O libraries=0A(such as HDF-5). We emphasize ways to use these interfaces that result in=0Ahigh performance, and benchmarks on real systems are used throughout to show=0Areal-world results.=0A=0AThis tutorial first discusses parallel file systems in detail (PFSs). We cover=0Ageneral concepts and examine four examples: GPFS, Lustre, PanFS, and=0APVFS. We examine the upper layers of the I/O stack, covering POSIX I/O,=0AMPI-IO, Parallel netCDF, and HDF5. We discuss interface features, show code=0Aexamples, and describe how application calls translate into PFS=0Aoperations. Finally we discuss I/O best practice. SUMMARY:Parallel IO In Practice PRIORITY:3 END:VEVENT END:VCALENDAR