Skip to content

Coll/han Improvements on algorithm selection through MCA and configuration file #10828

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

Merged
merged 1 commit into from
Oct 25, 2022
Merged

Coll/han Improvements on algorithm selection through MCA and configuration file #10828

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions ompi/mca/coll/han/Makefile.am
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
# of Tennessee Research Foundation. All rights
# reserved.
# Copyright (c) 2022 Amazon.com, Inc. or its affiliates. All Rights reserved.
# Copyright (c) 2022 BULL S.A.S. All rights reserved.
# $COPYRIGHT$
#
# Additional copyrights may follow
Expand All @@ -13,6 +14,7 @@
sources = \
coll_han.h \
coll_han_trigger.h \
coll_han_algorithms.h \
coll_han_dynamic.h \
coll_han_dynamic_file.h \
coll_han_barrier.c \
Expand All @@ -25,6 +27,7 @@ coll_han_allgather.c \
coll_han_component.c \
coll_han_module.c \
coll_han_trigger.c \
coll_han_algorithms.c \
coll_han_dynamic.c \
coll_han_dynamic_file.c \
coll_han_topo.c \
Expand Down
131 changes: 10 additions & 121 deletions ompi/mca/coll/han/coll_han.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,8 +2,8 @@
* Copyright (c) 2018-2020 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2020 Bull S.A.S. All rights reserved.
* Copyright (c) 2022 IBM Corporation. All rights reserved
* Copyright (c) 2020-2022 Bull S.A.S. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
Expand Down Expand Up @@ -40,6 +40,7 @@
#include "ompi/mca/coll/base/coll_base_functions.h"
#include "coll_han_trigger.h"
#include "ompi/mca/coll/han/coll_han_dynamic.h"
#include "coll_han_algorithms.h"

/*
* Today;
Expand Down Expand Up @@ -205,6 +206,7 @@ typedef struct mca_coll_han_component_t {
int han_priority;
/* whether output the log message */
int han_output;
int han_output_verbose; /* activation level of coll han verbosity */
/* segment size for bcast */
uint32_t han_bcast_segsize;
/* up level module for bcast */
Expand Down Expand Up @@ -242,6 +244,8 @@ typedef struct mca_coll_han_component_t {
*/
bool han_reproducible;
bool use_simple_algorithm[COLLCOUNT];
int use_algorithm[COLLCOUNT];
int use_algorithm_param[COLLCOUNT]; // MCA parmeter id for algo, to know if user provided

/* Dynamic configuration rules */
bool use_dynamic_file_rules;
Expand All @@ -250,7 +254,11 @@ typedef struct mca_coll_han_component_t {
/* Dynamic rules from file */
mca_coll_han_dynamic_rules_t dynamic_rules;
/* Dynamic rules from mca parameter */
COMPONENT_T mca_rules[COLLCOUNT][NB_TOPO_LVL];
COMPONENT_T mca_sub_components[COLLCOUNT][NB_TOPO_LVL];

int num_available_algorithms[COLLCOUNT]; // not counting "default" behaviour
/* to show algorithms in ompi_info */
mca_base_var_enum_value_t* algorithm_enumerator[COLLCOUNT];

/* Define maximum dynamic errors printed by rank 0 with a 0 verbosity level */
int max_dynamic_errors;
Expand Down Expand Up @@ -469,109 +477,7 @@ mca_coll_han_scatter_intra_dynamic(SCATTER_BASE_ARGS,

int mca_coll_han_barrier_intra_simple(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
/* Bcast */
int mca_coll_han_bcast_intra_simple(void *buff,
int count,
struct ompi_datatype_t *dtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int mca_coll_han_bcast_intra(void *buff, int count, struct ompi_datatype_t *dtype, int root,
struct ompi_communicator_t *comm, mca_coll_base_module_t * module);

/* Reduce */
int
mca_coll_han_reduce_intra_simple(const void *sbuf,
void* rbuf,
int count,
struct ompi_datatype_t *dtype,
ompi_op_t *op,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int
mca_coll_han_reduce_reproducible_decision(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int
mca_coll_han_reduce_reproducible(const void *sbuf,
void *rbuf,
int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);

int mca_coll_han_reduce_intra(const void *sbuf,
void *rbuf,
int count,
struct ompi_datatype_t *dtype,
ompi_op_t* op,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t * module);

/* Allreduce */
int
mca_coll_han_allreduce_intra_simple(const void *sbuf,
void *rbuf,
int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int
mca_coll_han_allreduce_reproducible_decision(struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
int
mca_coll_han_allreduce_reproducible(const void *sbuf,
void *rbuf,
int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);

int mca_coll_han_allreduce_intra(const void *sbuf,
void *rbuf,
int count,
struct ompi_datatype_t *dtype,
struct ompi_op_t *op,
struct ompi_communicator_t *comm, mca_coll_base_module_t * module);

/* Scatter */
int
mca_coll_han_scatter_intra(const void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm, mca_coll_base_module_t * module);
int
mca_coll_han_scatter_intra_simple(const void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t * module);

/* Gather */
int
mca_coll_han_gather_intra(const void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm, mca_coll_base_module_t * module);
int
mca_coll_han_gather_intra_simple(const void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
int root,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);
/* reordering after gather, for unordered ranks */
void
ompi_coll_han_reorder_gather(const void *sbuf,
Expand All @@ -580,21 +486,4 @@ ompi_coll_han_reorder_gather(const void *sbuf,
struct ompi_communicator_t *comm,
int * topo);



/* Allgather */
int
mca_coll_han_allgather_intra(const void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void *rbuf, int rcount,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm, mca_coll_base_module_t * module);
int
mca_coll_han_allgather_intra_simple(const void *sbuf, int scount,
struct ompi_datatype_t *sdtype,
void* rbuf, int rcount,
struct ompi_datatype_t *rdtype,
struct ompi_communicator_t *comm,
mca_coll_base_module_t *module);

#endif /* MCA_COLL_HAN_EXPORT_H */
Loading