Parallel I/O Benchmarks, Applications, Traces

Since I am often asked if I know of any parallel I/O benchmarks that are available, I have put together a list of the ones I am aware of (in no particular order). If you know of any benchmarks/applications that can be added to this list, let me (Rajeev Thakur) know.

Updated: July 18, 2014


  1. PnetCDF Benchmarks

  2. HACC I/O

  3. FLASH I/O
    An updated version for newer versions of parallel NetCDF and HDF-5 is available

  4. IOR

  5. Effective I/O Bandwidth Benchmark (b_eff_io)

  6. NAS BTIO Benchmark (see BT in NPB3.3-MPI)

  7. MPI Tile I/O

  8. FS_TEST (see this as a README)

  9. Trace-Driven Scientific I/O Benchmarks

  10. HPIO and S3asim

  11. Noncontig

  12. ioperf test in Silo

  13. Pattern I/O Benchmark (John Shalf, LBNL)

  14. MADbench2

  15. QIO

  16. PIO-Bench
    An article describing it is at

  17. Parkbench I/O Benchmarks


  1. Pagoda : Parallel Analysis of Geoscience Data

  2. Earth System Modeling Framework (ESMF)

  3. Nek5000

  4. NekCEM

  5. PISM: Parallel Ice Sheet Model

  6. MOAB


  8. Cactus

  9. mpiBLAST

  10. FLASH
    Also see FLASH I/O Benchmark above.

  11. Community Earth System Model (CESM)

  12. Regional Ocean Modeling System (ROMS)

  13. Global Cloud Resolving Model (GCRM)

  14. Weather Research & Forecasting Model (WRF)

  15. CM1

  16. Qbox

  17. GADGET

    NERSC web site

  19. MILC

  20. MPQC

  21. NaSt3DGP

  22. SIESTA

  23. CP2K

  24. LAMMPS


  1. Darshan

  2. PLFS

  3. LANL

  4. Sandia

  5. MAMBO