Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

#2281: Add new allreduce algorithms for Group, Objgroup and Collection #2337

Draft
wants to merge 70 commits into
base: develop
Choose a base branch
from
Draft

#2281: Add new allreduce algorithms for Group, Objgroup and Collection #2337

wants to merge 70 commits into from

Commits on Oct 10, 2024

  1. #2240: Initial work for new allreduce

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    e7e1a2e View commit details
    Browse the repository at this point in the history

  2. #2240: Semi working Rabenseifner

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    f90e5c7 View commit details
    Browse the repository at this point in the history

  3. #2240: Working Rabenseifner (non-commutative ops)

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    0715c52 View commit details
    Browse the repository at this point in the history

  4. #2240: Fix non power of 2 for new allreduce

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    0070e71 View commit details
    Browse the repository at this point in the history

  5. #2240: Initial work for adding recursive doubling allreduce algorithm

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    a2ad6e6 View commit details
    Browse the repository at this point in the history

  6. #2240: Make sure the order of reduce operations is correct

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    19df2f7 View commit details
    Browse the repository at this point in the history

  7. #2240: Working Recursive doubling

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    831168f View commit details
    Browse the repository at this point in the history

  8. #2240: Code cleanup and make Rabenseifner work with any Op type

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    594fce5 View commit details
    Browse the repository at this point in the history

  9. #2240: Improve accuracy of timing allreduce algorithms in allreduce.cc

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    b5095b6 View commit details
    Browse the repository at this point in the history

  10. #2240: Add unit tests for new allreduce and cleanup code

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    42cf608 View commit details
    Browse the repository at this point in the history

  11. #2240: DataHandler for Rabenseifner allreduce that provides common AP… 

    …I for various data types
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    0ee7848 View commit details
    Browse the repository at this point in the history

  12. #2240: Fix warnings

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    d70050e View commit details
    Browse the repository at this point in the history

  13. #2240: Update ObjGroup test to use custom DataHandler for Rabenseifne… 

    …r allreduce test
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    1f86e55 View commit details
    Browse the repository at this point in the history

  14. #2240: Add unit test for Rabenseifner with Kokkos::View as DataType a… 

    …nd fix compile issues realted to using Kokkos::View for allreduce
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    699a6d8 View commit details
    Browse the repository at this point in the history

  15. #2240: Move function definitions to impl.h file for Rabenseifner

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    5535449 View commit details
    Browse the repository at this point in the history

  16. #2240: Add allreduce print category and use it in rabenseifner instea… 

    …d of fmt::print
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    18d5090 View commit details
    Browse the repository at this point in the history

  17. #2240: Provide documentation for RecursiveDoubling algorithm

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    9d88016 View commit details
    Browse the repository at this point in the history

  18. #2240: Use vt_debug_print for RecursiveDoubling allreduce

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    5d79bd0 View commit details
    Browse the repository at this point in the history

  19. #2240: Update allreduce perf tests to use array of payload sizes

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    74bea76 View commit details
    Browse the repository at this point in the history

  20. #2240: Fix runtime failure in allreduce perf test

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    51ea74e View commit details
    Browse the repository at this point in the history

  21. #2240: Working allreduce perf test with Kokkos

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    7aeaa8e View commit details
    Browse the repository at this point in the history

  22. #2240: Working RecursiveDoubling with multiple allreduce in flight

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    593132f View commit details
    Browse the repository at this point in the history

  23. #2240: Update Rabenseifner to use ID for each allreduce and update tests

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    da46b94 View commit details
    Browse the repository at this point in the history

  24. #2240: Fix failing unit and performance tests for multiple allreduce … 

    …in flight
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    641c9dd View commit details
    Browse the repository at this point in the history

  25. #2240: Fix compile issues on some compilers and runtime issue with pa… 

    …ssing proxy by reference
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    1a0879e View commit details
    Browse the repository at this point in the history

  26. #2240: Update logs

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    3b9d697 View commit details
    Browse the repository at this point in the history

  27. #2240: Fix issues with handlers being executed and payload not being … 

    …initialized
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    27438ea View commit details
    Browse the repository at this point in the history

  28. #2240: Add helpers and use Kokkos::View for internals of Rabenseifner… 

    … when user's payload is View
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    6ba48d3 View commit details
    Browse the repository at this point in the history

  29. #2240: Store Reducers by tuple(ProxyType, DataType, OperandType)

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    150b883 View commit details
    Browse the repository at this point in the history

  30. #2281: Initial work to make collective group info contain the informa… 

    …tion about all the nodes from that group
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    b0a60b6 View commit details
    Browse the repository at this point in the history

  31. #2281: Working nodes information for each group node

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    d065c74 View commit details
    Browse the repository at this point in the history

  32. #2281: Working original Rabenseifner with groups

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    f68bc7f View commit details
    Browse the repository at this point in the history

  33. #2281: Working allreduce with Groups. Only thing missing is executing… 

    … final handler
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    172dc27 View commit details
    Browse the repository at this point in the history

  34. #2281: Working Handler execution on ObjGroup

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    d3280ae View commit details
    Browse the repository at this point in the history

  35. #2281: Fix issue with collective groups' default_group and root_node … 

    …information
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    eb9b71d View commit details
    Browse the repository at this point in the history

  36. #2281: Initial work for using new allreduce within collections

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    5cc8aa3 View commit details
    Browse the repository at this point in the history

  37. #2281: Working Rabenseifner (without final handler) for collection

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    4e29045 View commit details
    Browse the repository at this point in the history

  38. #2281: Add template param for Rabenseifner to determine how to procee… 

    …d with handler
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    26529f5 View commit details
    Browse the repository at this point in the history

  39. #2281: Working handler execution

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    a2245a1 View commit details
    Browse the repository at this point in the history

  40. #2281: Make final handler in Rabenseifner a callback

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    1d6800b View commit details
    Browse the repository at this point in the history

  41. #2281: Fix the issue with DataHan::reduce

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    db14b0a View commit details
    Browse the repository at this point in the history

  42. #2281: Make GroupManager and ObjManager also use callback for final h… 

    …andler (Rabenseifner)
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    fa93776 View commit details
    Browse the repository at this point in the history

  43. #2281: Working perf test with collection

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    503d64b View commit details
    Browse the repository at this point in the history

  44. #2281: Setup perf test for groups and remove the parent proxy, as we … 

    …now store the final handler with callback
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    5bbbb59 View commit details
    Browse the repository at this point in the history

  45. #2281: Remove RabenseifnerGroup as we've now integrated groups and co… 

    …llections into the original struct
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    23c0a6f View commit details
    Browse the repository at this point in the history

  46. #2281: Add unit tests for new callback (BcastCollective)

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    9f2cbaf View commit details
    Browse the repository at this point in the history

  47. #2281: Refactor Rabensifer a bit so it doesn't require that many temp… 

    …late arguments, also let the User specify which algorithm to use
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    57aa52b View commit details
    Browse the repository at this point in the history

  48. #2281: Implement state holder and use it for Rabenseifner

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    028a354 View commit details
    Browse the repository at this point in the history

  49. #2281: Implement clear function for StateHolder that frees dynamic me… 

    …mory allocated for allreduce state
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    1aa6f98 View commit details
    Browse the repository at this point in the history

  50. #2281: Working Rabenseifner with updated StateHolder

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    fde360f View commit details
    Browse the repository at this point in the history

  51. #2281: Typeless Rabensifner and small fixes to StateHolder

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    679344d View commit details
    Browse the repository at this point in the history

  52. #2281: Typeless RecursiveDoubling and general code refactor/cleanup

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    2e7b6e6 View commit details
    Browse the repository at this point in the history

  53. #2281: RecursiveDoubling allreduce - add Collection support

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    0c7f5dd View commit details
    Browse the repository at this point in the history

  54. #2281: Add unit tests for new Collection allreduce

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    c86244b View commit details
    Browse the repository at this point in the history

  55. #2281: Small fixes and code cleanup

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    76c377c View commit details
    Browse the repository at this point in the history

  56. #2281: Update documentation for new allreduce algorithms

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    1bafd3b View commit details
    Browse the repository at this point in the history

  57. #2281: Initial work for AllreduceHolder

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    30b1882 View commit details
    Browse the repository at this point in the history

  58. #2281: Fix runtime Kokkos when we run with both device/host memory sp… 

    …aces
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    0eb340b View commit details
    Browse the repository at this point in the history

  59. #2281: Working AllreduceHolder with all (Collection/Group/ObjGroup) c… 

    …omponents
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    3fb1ddd View commit details
    Browse the repository at this point in the history

  60. #2281: Fixed runtime issues with StateHolder generating extra ID

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    ae65729 View commit details
    Browse the repository at this point in the history

  61. #2281: Update license and add reduce_op.h file for Kokkos reduce oper… 

    …ators
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    d118fa9 View commit details
    Browse the repository at this point in the history

  62. #2281: Always create allreducers for new collective group

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    c61db0c View commit details
    Browse the repository at this point in the history

  63. #2281: New allreducers no longer require Objgroup for internal commun… 

    …ication. This helps for cases when not all Nodes are part of allreduce
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    7caed11 View commit details
    Browse the repository at this point in the history

  64. #2281: Code refactor and minor bug fixes

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    e24cacc View commit details
    Browse the repository at this point in the history

  65. #2281: Sort out the PendingSends in new allreduce methods

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    46c73af View commit details
    Browse the repository at this point in the history

  66. #2281: Don't use explicit Kokkos::HostSpace for Kokkos::View in Raben… 

    …seifnerMsg and ShouldUseView
    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    33c5e98 View commit details
    Browse the repository at this point in the history

  67. #2281: Cleanup StateHolder (move implementation to .cc impl.h files)

    @JacobDomagala
    JacobDomagala committed Oct 10, 2024
    Configuration menu
    Copy the full SHA
    ed1f880 View commit details
    Browse the repository at this point in the history

Commits on Oct 14, 2024

  1. #2281: Resolve issue with incorrect index generated by StateHolder::g… 

    …etNextID
    @JacobDomagala
    JacobDomagala committed Oct 14, 2024
    Configuration menu
    Copy the full SHA
    5c33e20 View commit details
    Browse the repository at this point in the history

Commits on Nov 5, 2024

  1. #2281: use or remove unused variables

    @cwschilly
    cwschilly committed Nov 5, 2024
    Configuration menu
    Copy the full SHA
    eee5e69 View commit details
    Browse the repository at this point in the history

Commits on Dec 3, 2024

  1. #2281: add perf tests for MPI_Allreduce

    @cwschilly
    cwschilly committed Dec 3, 2024
    Configuration menu
    Copy the full SHA
    4600310 View commit details
    Browse the repository at this point in the history