1 / 25

Flexible IO and Integration for Scientific Codes Through The Adaptable IO System (ADIOS)

Flexible IO and Integration for Scientific Codes Through The Adaptable IO System (ADIOS). Jay Lofstead (GT), Scott Klasky (ORNL), Karsten Schwan(GT), Norbert Podhorszki (ORNL), Chen Jin (ORNL) June 23, 2008. Outline. Motivation Design Goals Architecture Related Work Supported Features

gur
Download Presentation

Flexible IO and Integration for Scientific Codes Through The Adaptable IO System (ADIOS)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Flexible IO and Integration for Scientific CodesThrough The Adaptable IO System (ADIOS) Jay Lofstead (GT), Scott Klasky (ORNL), Karsten Schwan(GT), Norbert Podhorszki (ORNL), Chen Jin (ORNL) June 23, 2008 Jay Lofstead

  2. Outline • Motivation • Design Goals • Architecture • Related Work • Supported Features • Examples • Codes & Performance • Future Work • Collaborators Jay Lofstead

  3. Motivation • Multiple HPC architectures • Cray, IB-based clusters, BlueGene • Many different APIs • MPI-IO, POSIX, HDF5, netCDF • GTC (fusion) has changed IO routines 8 times so far based on moving platforms • Different IO patterns • Restarts, analysis, diagnostics • Different combinations provide different levels of I/O performance Jay Lofstead

  4. Design Goals • ADIOS Fortran and C based API almost as simple as standard POSIX IO • External configuration to describe metadata and control IO settings • Take advantage of existing IO techniques (no new native IO methods) Fast, simple-to-write, efficient IO for multiple platforms without changing the source code Jay Lofstead

  5. Design Goals ADIOS Contributions: • Data Grouping: Multiple, independently controlled IO settings • diagnostics, restarts, analysis, viz • Optional data items: Different data from different processes • Single group write has header from a single proc and data from all • Data sizes change dynamically: Datasets vary in size • Run size changes data sizes; also AMR codes • Constructed output is special: Reused buffers must be handled properly • stack temporaries... • IO memory is second to science: Buffer space for IO is strictly limited • respect the memory needs of the scientific codes Jay Lofstead

  6. Design Goals Provide support for these future features: • Provide integration opportunities for other systems • Workflow • In situ visualization • Other simulations • Provenance collection • Other data management tools • Science aware data compression Jay Lofstead

  7. Architecture Scientific Codes • Thin API • XML file • data groupings with annotation • IO method selection • buffer sizes • Common tools • Buffering • Scheduling • Pluggable IO routines External Metadata (XML file) ADIOS API buffering schedule feedback DART HDF-5 MPI-IO pnetCDF POSIX IO Viz Engines LIVE/DataTap Others (plug-in) Jay Lofstead

  8. Architecture • Data groupings • logical groups of related items written at the same time. • Not necessarily one group per writing event • IO Methods • Choose what works best for each grouping • Vetted and/or written by experts for each • POSIX (Wei-keng Lao Northwestern) • MPI-IO (Steve Hodson, ORNL) • MPI-IO Collective (Wei-keng Lao Northwestern) • NULL (Jay Lofstead GT) • Ga Tech DataTap Asynchronous (Hasan Abbasi GT) • others.. Jay Lofstead

  9. Related Work • Specialty APIs • HDF-5 – complex API • Parallel netCDF – no structure • File system aware middleware • MPI ADIO layer – File system connection, complex API • Parallel File systems • Lustre – Metadata server issues • PVFS2 – client complexity • LWFS – client complexity • GPFS, pNFS, Panasas Jay Lofstead

  10. Supported Features • Platforms tested • Cray CNL (ORNL Jaguar) • Cray Catamount (ORNL old-Jaguar and SNL Redstorm) • Linux Infiniband (ORNL Ewok) • IO Methods • MPI-IO independent, MPI-IO collective, POSIX, NULL, Ga Tech DataTap asynchronous, Rutgers DART asynchronous Jay Lofstead

  11. Examples • Setup/Cleanup code call adios_init (‘config.xml’) ... call adios_finalize (my_rank) • adios_init – reads the XML file (once for every process) • adios_finalize – provide opportunity for cleanup Jay Lofstead

  12. Examples • IO code call adios_open (handle, ‘filename’, ‘groupname’) call adios_write (handle, ‘varname’, var) ADIOS_WRITE(handle,var_name) ... call adios_close (handle) Jay Lofstead

  13. Examples • IO code call adios_open (handle, ‘filename’, ‘groupname’) ADIOS_GWRITE (handle) call adios_close (handle) - ADIOS_GREAD coming soon Jay Lofstead

  14. Examples • IO hints (for asynchronous IO) • call adios_end_iteration () • pacing hints for asynchronous transport methods • use in conjunction with ‘iterations’ attribute of method element in XML • call adios_begin_calculation () • hint for asynchronous IO that a low-IO phase is starting • call adios_end_calculation () • hint for asynchronous IO that a low-IO phase is ending Jay Lofstead

  15. Examples • XML file contents (data elements) <adios-confighost-language=“Fortran”> <adios-groupname=“restart” coordination-communicator=“comm” coordination-var=“cvar”> <varname=“elements” type=“integer” write=“no”/> <varname=“data” type=“double” path=“/” dimensions=“elements” copy-on-write=“yes”/> </adios-group> </adios-config> Jay Lofstead

  16. Examples • XML file contents (other) <attributename=“description” path=“/data” value=“simulation particle data”/> <global-boundsdimensions=“..” offsets=“..”> <var .../> </global-bounds> <mesh ..../> <methodmethod=“DataTap” iterations=“50” group=“restart”>params</method> <methodmethod=“visit” group=“restart”/> <buffersize-MB=“100” allocate-time=“now”/> Jay Lofstead

  17. Codes & Performance • Codes • GTC (fusion), Chimera (supernova), GTS (fusion), XGC0 (fusion), XGC1 (fusion), Flash (astrophysics), S3D (combustion) • Performance Measurements • GTC: nearly 25 GB/sec • GTS: 12 GB/sec • Chimera: 6.5% reduction in wall clock time Jay Lofstead

  18. Codes & Performance Jay Lofstead

  19. Codes & Performance 12 GB/restart Old GTC IO GTC On Rizzo at ORNL New GT ADIOS/DataTap IO Jay Lofstead

  20. Codes & Performance • June 7, 2008: 24 hour GTC run on Jaguar at ORNL • 93% of machine (28,672 cores) • MPI-OpenMP mixed model on quad-core nodes (7168 MPI procs) • three interruptions total (simple node failure) with 2 10+ hour runs • Wrote 56 TB of data at 20 GB/sec (21 TB for post analysis) • IO overhead 3% of wall clock time • Mixed IO methods of synchronous MPI-IO and POSIX IO configured in the XML file Jay Lofstead

  21. Support Tools • bpdump – similar to h5dump • bp2h5 – convert output file to HDF5 • bp2nc – convert output to netCDF Jay Lofstead

  22. Future Work • Work on read functionality • Index files • Additional native transports • direct parallel HDF5 and parallel netCDF • asynchronous MPI • Visit • others as demanded Jay Lofstead

  23. Future Work • Offload to staging area for other processing Jay Lofstead

  24. Collaborators • Chen Jin, Scott Klasky, Stephen Hodson, James B. White III, Weikuan Yu (Oak Ridge National Laboratory) • Hasan Abbasi, Greg Eisenhauer, Karsten Schwan, Matthew Wolf, Fang Zheng (Georgia Tech) • Wei-keng Liao, Alok Choudhary, Northwestern University • Manish Parashar, Ciprian Docan, Rutgers University. • Ron Oldfield, Sandia Labs Jay Lofstead

  25. References • Jay Lofstead, Scott Klasky, Karsten Schwan, Norbert Podhorszki, Chen Jin. “Flexible IO and Integration for Scientific Codes Through The Adaptable IO System (ADIOS).” In CLADE 2008 at HPDC, Boston, MA, June 2008. ACM. • Hasan Abbasi, Matthew Wolf, Karsten Schwan. “Live data workspace: A flexible, dynamic and extensible platform for petascale applications.” In Cluster Computing, Austin, TX, September 2007. IEEE International. • Chen Jin, Scott Klasky, Stephen Hodson, Weikuan Yu (Oak Ridge National Laboratory) Jay Lofstead, Hasan Abbasi, Karsten Schwan, Matthew Wolf (Georgia Tech) Wei-keng Liao, Alok Choudhary(Northwestern University) Manish Parashar, Ciprian Docan, Rutgers University. Ron Oldfield (Sandia National Laboratories). “Adaptive IO System (ADIOS).” In CUG 2008. Helsinki, Findland, May 2008. • Sarp Oral. NCCS Jaguar GTC 128 Nodes Test Results. http://users.nccs.gov/~oral/jagregtests/gtc128.html. Jay Lofstead

More Related