2013-01-27 23:25:10 +00:00
|
|
|
/*
|
2013-02-21 18:19:46 +00:00
|
|
|
* Copyright (c) 2012-2013 Los Alamos National Security, LLC.
|
2013-01-27 23:25:10 +00:00
|
|
|
* All rights reserved.
|
For large scale systems, we would like to avoid doing a full modex during MPI_Init so that launch will scale a little better. At the moment, our options are somewhat limited as only a few BTLs don't immediately call modex_recv on all procs during startup. However, for those situations where someone can take advantage of it, add the ability to do a "modex on demand" retrieval of data from remote procs when we launch via mpirun.
NOTE: launch performance will be absolutely awful if you do this with BTLs that aren't configured to modex_recv on first message!
Even with "modex on demand", we still have to do a barrier in place of the modex - we simply don't move any data around, which does reduce the time impact. The barrier is required to ensure that the other proc has in fact registered all its BTL info and therefore is prepared to hand over a complete data package. Otherwise, you may not get the info you need. In addition, the shared memory BTL can fail to properly rendezvous as it expects the barrier to be in place.
This behavior will *only* take effect under the following conditions:
1. launched via mpirun
2. #procs is greater than ompi_hostname_cutoff, which defaults to UINT32_MAX
3. mca param rte_orte_direct_modex is set to 1. At the moment, we are having problems getting this param to register properly, so only the first two conditions are in effect. Still, the bottom line is you have to *want* this behavior to get it.
The planned next evolution of this will be to make the direct modex be non-blocking - this will require two fixes:
1. if the remote proc doesn't have the required info, then let it delay its response until it does. This means we need a way for the MPI layer to tell the RTE "I am done entering modex data".
2. adjust the SM rendezvous logic to loop until the required file has been created
Creating a placeholder to bring this over to 1.7.5 when ready.
cmr=v1.7.5:reviewer=hjelmn:subject=Enable direct modex at scale
This commit was SVN r30259.
2014-01-11 17:36:06 +00:00
|
|
|
* Copyright (c) 2013-2014 Intel, Inc. All rights reserved
|
2013-01-27 23:25:10 +00:00
|
|
|
*
|
|
|
|
|
* $COPYRIGHT$
|
|
|
|
|
*
|
|
|
|
|
* Additional copyrights may follow
|
|
|
|
|
*
|
|
|
|
|
* $HEADER$
|
|
|
|
|
*
|
|
|
|
|
* When this component is used, this file is included in the rest of
|
|
|
|
|
* the OPAL/ORTE/OMPI code base via ompi/mca/rte/rte.h. As such,
|
|
|
|
|
* this header represents the public interface to this static component.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#ifndef MCA_OMPI_RTE_ORTE_H
|
|
|
|
|
#define MCA_OMPI_RTE_ORTE_H
|
|
|
|
|
|
|
|
|
|
#include "ompi_config.h"
|
|
|
|
|
#include "ompi/constants.h"
|
|
|
|
|
|
|
|
|
|
#include "ompi/info/info.h"
|
2013-12-17 03:26:00 +00:00
|
|
|
struct ompi_proc_t;
|
2013-01-27 23:25:10 +00:00
|
|
|
|
|
|
|
|
#include "orte/types.h"
|
|
|
|
|
#include "orte/mca/errmgr/errmgr.h"
|
|
|
|
|
#include "orte/mca/grpcomm/grpcomm.h"
|
|
|
|
|
#include "orte/mca/rml/base/rml_contact.h"
|
|
|
|
|
#include "orte/mca/rml/rml.h"
|
|
|
|
|
#include "orte/mca/routed/routed.h"
|
|
|
|
|
#include "orte/runtime/orte_data_server.h"
|
|
|
|
|
#include "orte/runtime/runtime.h"
|
|
|
|
|
#include "orte/util/name_fns.h"
|
|
|
|
|
#include "orte/util/proc_info.h"
|
|
|
|
|
|
|
|
|
|
BEGIN_C_DECLS
|
|
|
|
|
|
|
|
|
|
/* Process name objects and operations */
|
|
|
|
|
typedef orte_process_name_t ompi_process_name_t;
|
|
|
|
|
typedef orte_jobid_t ompi_jobid_t;
|
|
|
|
|
typedef orte_vpid_t ompi_vpid_t;
|
|
|
|
|
typedef orte_ns_cmp_bitmask_t ompi_rte_cmp_bitmask_t;
|
|
|
|
|
#define OMPI_PROC_MY_NAME ORTE_PROC_MY_NAME
|
|
|
|
|
#define OMPI_NAME_PRINT(a) ORTE_NAME_PRINT(a)
|
|
|
|
|
#define ompi_rte_compare_name_fields(a, b, c) orte_util_compare_name_fields(a, b, c)
|
|
|
|
|
#define OMPI_NAME_WILDCARD ORTE_NAME_WILDCARD
|
|
|
|
|
#define OMPI_NODE_RANK_INVALID ORTE_NODE_RANK_INVALID
|
2013-02-21 18:19:46 +00:00
|
|
|
#define OMPI_LOCAL_RANK_INVALID ORTE_LOCAL_RANK_INVALID
|
2013-01-27 23:25:10 +00:00
|
|
|
#define OMPI_RTE_CMP_JOBID ORTE_NS_CMP_JOBID
|
|
|
|
|
#define OMPI_RTE_CMP_VPID ORTE_NS_CMP_VPID
|
|
|
|
|
#define OMPI_RTE_CMP_ALL ORTE_NS_CMP_ALL
|
|
|
|
|
#define ompi_rte_hash_name(a) orte_util_hash_name(a)
|
|
|
|
|
/* This is the DSS tag to serialize a proc name */
|
|
|
|
|
#define OMPI_NAME ORTE_NAME
|
|
|
|
|
#define OMPI_PROCESS_NAME_HTON ORTE_PROCESS_NAME_HTON
|
|
|
|
|
#define OMPI_PROCESS_NAME_NTOH ORTE_PROCESS_NAME_NTOH
|
2013-09-27 00:37:49 +00:00
|
|
|
#define OMPI_RTE_NODE_ID ORTE_DB_DAEMON_VPID
|
|
|
|
|
#define OMPI_RTE_MY_NODEID ORTE_PROC_MY_DAEMON->vpid
|
****************************************************************
This change contains a non-mandatory modification
of the MPI-RTE interface. Anyone wishing to support
coprocessors such as the Xeon Phi may wish to add
the required definition and underlying support
****************************************************************
Add locality support for coprocessors such as the Intel Xeon Phi.
Detecting that we are on a coprocessor inside of a host node isn't straightforward. There are no good "hooks" provided for programmatically detecting that "we are on a coprocessor running its own OS", and the ORTE daemon just thinks it is on another node. However, in order to properly use the Phi's public interface for MPI transport, it is necessary that the daemon detect that it is colocated with procs on the host.
So we have to split the locality to separately record "on the same host" vs "on the same board". We already have the board-level locality flag, but not quite enough flexibility to handle this use-case. Thus, do the following:
1. add OPAL_PROC_ON_HOST flag to indicate we share a host, but not necessarily the same board
2. modify OPAL_PROC_ON_NODE to indicate we share both a host AND the same board. Note that we have to modify the OPAL_PROC_ON_LOCAL_NODE macro to explicitly check both conditions
3. add support in opal/mca/hwloc/base/hwloc_base_util.c for the host to check for coprocessors, and for daemons to check to see if they are on a coprocessor. The former is done via hwloc, but support for the latter is not yet provided by hwloc. So the code for detecting we are on a coprocessor currently is Xeon Phi specific - hopefully, we will find more generic methods in the future.
4. modify the orted and the hnp startup so they check for coprocessors and to see if they are on a coprocessor, and have the orteds pass that info back in their callback message. Automatically detect that coprocessors have been found and identify which coprocessors are on which hosts. Note that this algo isn't scalable at the moment - this will hopefully be improved over time.
5. modify the ompi proc locality detection function to look for coprocessor host info IF the OMPI_RTE_HOST_ID database key has been defined. RTE's that choose not to provide this support do not have to do anything - the associated code will simply be ignored.
6. include some cleanup of the hwloc open/close code so it conforms to how we did things in other frameworks (e.g., having a single "frame" file instead of open/close). Also, fix the locality flags - e.g., being on the same node means you must also be on the same cluster/cu, so ensure those flags are also set.
cmr:v1.7.4:reviewer=hjelmn
This commit was SVN r29435.
2013-10-14 16:52:58 +00:00
|
|
|
#define OMPI_RTE_HOST_ID ORTE_DB_HOSTID
|
2013-01-27 23:25:10 +00:00
|
|
|
|
|
|
|
|
/* Collective objects and operations */
|
|
|
|
|
#define ompi_rte_collective_t orte_grpcomm_collective_t
|
|
|
|
|
typedef orte_grpcomm_coll_id_t ompi_rte_collective_id_t;
|
For large scale systems, we would like to avoid doing a full modex during MPI_Init so that launch will scale a little better. At the moment, our options are somewhat limited as only a few BTLs don't immediately call modex_recv on all procs during startup. However, for those situations where someone can take advantage of it, add the ability to do a "modex on demand" retrieval of data from remote procs when we launch via mpirun.
NOTE: launch performance will be absolutely awful if you do this with BTLs that aren't configured to modex_recv on first message!
Even with "modex on demand", we still have to do a barrier in place of the modex - we simply don't move any data around, which does reduce the time impact. The barrier is required to ensure that the other proc has in fact registered all its BTL info and therefore is prepared to hand over a complete data package. Otherwise, you may not get the info you need. In addition, the shared memory BTL can fail to properly rendezvous as it expects the barrier to be in place.
This behavior will *only* take effect under the following conditions:
1. launched via mpirun
2. #procs is greater than ompi_hostname_cutoff, which defaults to UINT32_MAX
3. mca param rte_orte_direct_modex is set to 1. At the moment, we are having problems getting this param to register properly, so only the first two conditions are in effect. Still, the bottom line is you have to *want* this behavior to get it.
The planned next evolution of this will be to make the direct modex be non-blocking - this will require two fixes:
1. if the remote proc doesn't have the required info, then let it delay its response until it does. This means we need a way for the MPI layer to tell the RTE "I am done entering modex data".
2. adjust the SM rendezvous logic to loop until the required file has been created
Creating a placeholder to bring this over to 1.7.5 when ready.
cmr=v1.7.5:reviewer=hjelmn:subject=Enable direct modex at scale
This commit was SVN r30259.
2014-01-11 17:36:06 +00:00
|
|
|
OMPI_DECLSPEC int ompi_rte_modex(ompi_rte_collective_t *coll);
|
2013-01-27 23:25:10 +00:00
|
|
|
#define ompi_rte_barrier(a) orte_grpcomm.barrier(a)
|
|
|
|
|
|
|
|
|
|
/* Process info struct and values */
|
|
|
|
|
typedef orte_node_rank_t ompi_node_rank_t;
|
2013-02-22 17:48:53 +00:00
|
|
|
typedef orte_local_rank_t ompi_local_rank_t;
|
2013-01-27 23:25:10 +00:00
|
|
|
#define ompi_process_info orte_process_info
|
|
|
|
|
#define ompi_rte_proc_is_bound orte_proc_is_bound
|
|
|
|
|
|
|
|
|
|
/* Error handling objects and operations */
|
|
|
|
|
OMPI_DECLSPEC void ompi_rte_abort(int error_code, char *fmt, ...);
|
2013-10-08 18:37:59 +00:00
|
|
|
#define ompi_rte_abort_peers(a, b, c) orte_errmgr.abort_peers(a, b, c)
|
Per the RFC and discussion on the devel list, update the RTE-MPI error handling interface. There are a few differences in the code from the original RFC that came out of the discussion - I've captured those in the following writeup
George and I were talking about ORTE's error handling the other day in regards to the right way to deal with errors in the updated OOB. Specifically, it seemed a bad idea for a library such as ORTE to be aborting the job on its own prerogative. If we lose a connection or cannot send a message, then we really should just report it upwards and let the application and/or upper layers decide what to do about it.
The current code base only allows a single error callback to exist, which seemed unduly limiting. So, based on the conversation, I've modified the errmgr interface to provide a mechanism for registering any number of error handlers (this replaces the current "set_fault_callback" API). When an error occurs, these handlers will be called in order until one responds that the error has been "resolved" - i.e., no further action is required - by returning OMPI_SUCCESS. The default MPI layer error handler is specified to go "last" and calls mpi_abort, so the current "abort" behavior is preserved unless other error handlers are registered.
In the register_callback function, I provide an "order" param so you can specify "this callback must come first" or "this callback must come last". Seemed to me that we will probably have different code areas registering callbacks, and one might require it go first (the default "abort" will always require it go last). So you can append and prepend, or go first. Note that only one registration can declare itself "first" or "last", and since the default "abort" callback automatically takes "last", that one isn't available. :-)
The errhandler callback function passes an opal_pointer_array of structs, each of which contains the name of the proc involved (which can be yourself for internal errors) and the error code. This is a change from the current fault callback which returned an opal_pointer_array of just process names. Rationale is that you might need to see the cause of the error to decide what action to take. I realize that isn't a requirement for remote procs, but remember that we will use the SAME interface to report RTE errors internal to the proc itself. In those cases, you really do need to see the error code. It is legal to pass a NULL for the pointer array (e.g., when reporting an internal failure without error code), so handlers must be prepared for that possibility. If people find that too burdensome, we can remove it.
Should we ever decide to create a separate callback path for internal errors vs remote process failures, or if we decide to do something different based on experience, then we can adjust this API.
This commit was SVN r28852.
2013-07-19 01:08:53 +00:00
|
|
|
#define OMPI_RTE_ERRHANDLER_FIRST ORTE_ERRMGR_CALLBACK_FIRST
|
|
|
|
|
#define OMPI_RTE_ERRHANDLER_LAST ORTE_ERRMGR_CALLBACK_LAST
|
|
|
|
|
#define OMPI_RTE_ERRHANDLER_PREPEND ORTE_ERRMGR_CALLBACK_PREPEND
|
|
|
|
|
#define OMPI_RTE_ERRHANDLER_APPEND ORTE_ERRMGR_CALLBACK_APPEND
|
|
|
|
|
typedef orte_error_t ompi_rte_error_report_t;
|
|
|
|
|
#define ompi_rte_register_errhandler(a, b) orte_errmgr.register_error_callback(a, b)
|
2013-01-27 23:25:10 +00:00
|
|
|
#define OMPI_ERROR_LOG ORTE_ERROR_LOG
|
|
|
|
|
|
|
|
|
|
/* Init and finalize objects and operations */
|
|
|
|
|
#define ompi_rte_init(a, b) orte_init(a, b, ORTE_PROC_MPI)
|
|
|
|
|
#define ompi_rte_finalize() orte_finalize()
|
|
|
|
|
OMPI_DECLSPEC void ompi_rte_wait_for_debugger(void);
|
|
|
|
|
|
|
|
|
|
/* Database operations */
|
Per the meeting on moving the BTLs to OPAL, move the ORTE database "db" framework to OPAL so the relocated BTLs can access it. Because the data is indexed by process, this requires that we define a new "opal_identifier_t" that corresponds to the orte_process_name_t struct. In order to support multiple run-times, this is defined in opal/mca/db/db_types.h as a uint64_t without identifying the meaning of any part of that data.
A few changes were required to support this move:
1. the PMI component used to identify rte-related data (e.g., host name, bind level) and package them as a unit to reduce the number of PMI keys. This code was moved up to the ORTE layer as the OPAL layer has no understanding of these concepts. In addition, the component locally stored data based on process jobid/vpid - this could no longer be supported (see below for the solution).
2. the hash component was updated to use the new opal_identifier_t instead of orte_process_name_t as its index for storing data in the hash tables. Previously, we did a hash on the vpid and stored the data in a 32-bit hash table. In the revised system, we don't see a separate "vpid" field - we only have a 64-bit opaque value. The orte_process_name_t hash turned out to do nothing useful, so we now store the data in a 64-bit hash table. Preliminary tests didn't show any identifiable change in behavior or performance, but we'll have to see if a move back to the 32-bit table is required at some later time.
3. the db framework was a "select one" system. However, since the PMI component could no longer use its internal storage system, the framework has now been changed to a "select many" mode of operation. This allows the hash component to handle all internal storage, while the PMI component only handles pushing/pulling things from the PMI system. This was something we had planned for some time - when fetching data, we first check internal storage to see if we already have it, and then automatically go to the global system to look for it if we don't. Accordingly, the framework was provided with a custom query function used during "select" that lets you seperately specify the "store" and "fetch" ordering.
4. the ORTE grpcomm and ess/pmi components, and the nidmap code, were updated to work with the new db framework and to specify internal/global storage options.
No changes were made to the MPI layer, except for modifying the ORTE component of the OMPI/rte framework to support the new db framework.
This commit was SVN r28112.
2013-02-26 17:50:04 +00:00
|
|
|
OMPI_DECLSPEC int ompi_rte_db_store(const ompi_process_name_t *nm, const char* key,
|
|
|
|
|
const void *data, opal_data_type_t type);
|
2013-12-17 03:26:00 +00:00
|
|
|
OMPI_DECLSPEC int ompi_rte_db_fetch(const struct ompi_proc_t *proc,
|
Per the meeting on moving the BTLs to OPAL, move the ORTE database "db" framework to OPAL so the relocated BTLs can access it. Because the data is indexed by process, this requires that we define a new "opal_identifier_t" that corresponds to the orte_process_name_t struct. In order to support multiple run-times, this is defined in opal/mca/db/db_types.h as a uint64_t without identifying the meaning of any part of that data.
A few changes were required to support this move:
1. the PMI component used to identify rte-related data (e.g., host name, bind level) and package them as a unit to reduce the number of PMI keys. This code was moved up to the ORTE layer as the OPAL layer has no understanding of these concepts. In addition, the component locally stored data based on process jobid/vpid - this could no longer be supported (see below for the solution).
2. the hash component was updated to use the new opal_identifier_t instead of orte_process_name_t as its index for storing data in the hash tables. Previously, we did a hash on the vpid and stored the data in a 32-bit hash table. In the revised system, we don't see a separate "vpid" field - we only have a 64-bit opaque value. The orte_process_name_t hash turned out to do nothing useful, so we now store the data in a 64-bit hash table. Preliminary tests didn't show any identifiable change in behavior or performance, but we'll have to see if a move back to the 32-bit table is required at some later time.
3. the db framework was a "select one" system. However, since the PMI component could no longer use its internal storage system, the framework has now been changed to a "select many" mode of operation. This allows the hash component to handle all internal storage, while the PMI component only handles pushing/pulling things from the PMI system. This was something we had planned for some time - when fetching data, we first check internal storage to see if we already have it, and then automatically go to the global system to look for it if we don't. Accordingly, the framework was provided with a custom query function used during "select" that lets you seperately specify the "store" and "fetch" ordering.
4. the ORTE grpcomm and ess/pmi components, and the nidmap code, were updated to work with the new db framework and to specify internal/global storage options.
No changes were made to the MPI layer, except for modifying the ORTE component of the OMPI/rte framework to support the new db framework.
This commit was SVN r28112.
2013-02-26 17:50:04 +00:00
|
|
|
const char *key,
|
|
|
|
|
void **data, opal_data_type_t type);
|
2013-12-17 03:26:00 +00:00
|
|
|
OMPI_DECLSPEC int ompi_rte_db_fetch_pointer(const struct ompi_proc_t *proc,
|
Per the meeting on moving the BTLs to OPAL, move the ORTE database "db" framework to OPAL so the relocated BTLs can access it. Because the data is indexed by process, this requires that we define a new "opal_identifier_t" that corresponds to the orte_process_name_t struct. In order to support multiple run-times, this is defined in opal/mca/db/db_types.h as a uint64_t without identifying the meaning of any part of that data.
A few changes were required to support this move:
1. the PMI component used to identify rte-related data (e.g., host name, bind level) and package them as a unit to reduce the number of PMI keys. This code was moved up to the ORTE layer as the OPAL layer has no understanding of these concepts. In addition, the component locally stored data based on process jobid/vpid - this could no longer be supported (see below for the solution).
2. the hash component was updated to use the new opal_identifier_t instead of orte_process_name_t as its index for storing data in the hash tables. Previously, we did a hash on the vpid and stored the data in a 32-bit hash table. In the revised system, we don't see a separate "vpid" field - we only have a 64-bit opaque value. The orte_process_name_t hash turned out to do nothing useful, so we now store the data in a 64-bit hash table. Preliminary tests didn't show any identifiable change in behavior or performance, but we'll have to see if a move back to the 32-bit table is required at some later time.
3. the db framework was a "select one" system. However, since the PMI component could no longer use its internal storage system, the framework has now been changed to a "select many" mode of operation. This allows the hash component to handle all internal storage, while the PMI component only handles pushing/pulling things from the PMI system. This was something we had planned for some time - when fetching data, we first check internal storage to see if we already have it, and then automatically go to the global system to look for it if we don't. Accordingly, the framework was provided with a custom query function used during "select" that lets you seperately specify the "store" and "fetch" ordering.
4. the ORTE grpcomm and ess/pmi components, and the nidmap code, were updated to work with the new db framework and to specify internal/global storage options.
No changes were made to the MPI layer, except for modifying the ORTE component of the OMPI/rte framework to support the new db framework.
This commit was SVN r28112.
2013-02-26 17:50:04 +00:00
|
|
|
const char *key,
|
|
|
|
|
void **data, opal_data_type_t type);
|
2013-12-17 03:26:00 +00:00
|
|
|
OMPI_DECLSPEC int ompi_rte_db_fetch_multiple(const struct ompi_proc_t *proc,
|
Per the meeting on moving the BTLs to OPAL, move the ORTE database "db" framework to OPAL so the relocated BTLs can access it. Because the data is indexed by process, this requires that we define a new "opal_identifier_t" that corresponds to the orte_process_name_t struct. In order to support multiple run-times, this is defined in opal/mca/db/db_types.h as a uint64_t without identifying the meaning of any part of that data.
A few changes were required to support this move:
1. the PMI component used to identify rte-related data (e.g., host name, bind level) and package them as a unit to reduce the number of PMI keys. This code was moved up to the ORTE layer as the OPAL layer has no understanding of these concepts. In addition, the component locally stored data based on process jobid/vpid - this could no longer be supported (see below for the solution).
2. the hash component was updated to use the new opal_identifier_t instead of orte_process_name_t as its index for storing data in the hash tables. Previously, we did a hash on the vpid and stored the data in a 32-bit hash table. In the revised system, we don't see a separate "vpid" field - we only have a 64-bit opaque value. The orte_process_name_t hash turned out to do nothing useful, so we now store the data in a 64-bit hash table. Preliminary tests didn't show any identifiable change in behavior or performance, but we'll have to see if a move back to the 32-bit table is required at some later time.
3. the db framework was a "select one" system. However, since the PMI component could no longer use its internal storage system, the framework has now been changed to a "select many" mode of operation. This allows the hash component to handle all internal storage, while the PMI component only handles pushing/pulling things from the PMI system. This was something we had planned for some time - when fetching data, we first check internal storage to see if we already have it, and then automatically go to the global system to look for it if we don't. Accordingly, the framework was provided with a custom query function used during "select" that lets you seperately specify the "store" and "fetch" ordering.
4. the ORTE grpcomm and ess/pmi components, and the nidmap code, were updated to work with the new db framework and to specify internal/global storage options.
No changes were made to the MPI layer, except for modifying the ORTE component of the OMPI/rte framework to support the new db framework.
This commit was SVN r28112.
2013-02-26 17:50:04 +00:00
|
|
|
const char *key,
|
|
|
|
|
opal_list_t *kvs);
|
|
|
|
|
OMPI_DECLSPEC int ompi_rte_db_remove(const ompi_process_name_t *nm,
|
|
|
|
|
const char *key);
|
2013-01-27 23:25:10 +00:00
|
|
|
#define OMPI_DB_HOSTNAME ORTE_DB_HOSTNAME
|
|
|
|
|
#define OMPI_DB_LOCALITY ORTE_DB_LOCALITY
|
2013-11-14 17:01:43 +00:00
|
|
|
#define OMPI_DB_GLOBAL_RANK ORTE_DB_GLOBAL_RANK
|
2013-01-27 23:25:10 +00:00
|
|
|
|
|
|
|
|
/* Communications */
|
|
|
|
|
typedef orte_rml_tag_t ompi_rml_tag_t;
|
As per the RFC, bring in the ORTE async progress code and the rewrite of OOB:
*** THIS RFC INCLUDES A MINOR CHANGE TO THE MPI-RTE INTERFACE ***
Note: during the course of this work, it was necessary to completely separate the MPI and RTE progress engines. There were multiple places in the MPI layer where ORTE_WAIT_FOR_COMPLETION was being used. A new OMPI_WAIT_FOR_COMPLETION macro was created (defined in ompi/mca/rte/rte.h) that simply cycles across opal_progress until the provided flag becomes false. Places where the MPI layer blocked waiting for RTE to complete an event have been modified to use this macro.
***************************************************************************************
I am reissuing this RFC because of the time that has passed since its original release. Since its initial release and review, I have debugged it further to ensure it fully supports tests like loop_spawn. It therefore seems ready for merge back to the trunk. Given its prior review, I have set the timeout for one week.
The code is in https://bitbucket.org/rhc/ompi-oob2
WHAT: Rewrite of ORTE OOB
WHY: Support asynchronous progress and a host of other features
WHEN: Wed, August 21
SYNOPSIS:
The current OOB has served us well, but a number of limitations have been identified over the years. Specifically:
* it is only progressed when called via opal_progress, which can lead to hangs or recursive calls into libevent (which is not supported by that code)
* we've had issues when multiple NICs are available as the code doesn't "shift" messages between transports - thus, all nodes had to be available via the same TCP interface.
* the OOB "unloads" incoming opal_buffer_t objects during the transmission, thus preventing use of OBJ_RETAIN in the code when repeatedly sending the same message to multiple recipients
* there is no failover mechanism across NICs - if the selected NIC (or its attached switch) fails, we are forced to abort
* only one transport (i.e., component) can be "active"
The revised OOB resolves these problems:
* async progress is used for all application processes, with the progress thread blocking in the event library
* each available TCP NIC is supported by its own TCP module. The ability to asynchronously progress each module independently is provided, but not enabled by default (a runtime MCA parameter turns it "on")
* multi-address TCP NICs (e.g., a NIC with both an IPv4 and IPv6 address, or with virtual interfaces) are supported - reachability is determined by comparing the contact info for a peer against all addresses within the range covered by the address/mask pairs for the NIC.
* a message that arrives on one TCP NIC is automatically shifted to whatever NIC that is connected to the next "hop" if that peer cannot be reached by the incoming NIC. If no TCP module will reach the peer, then the OOB attempts to send the message via all other available components - if none can reach the peer, then an "error" is reported back to the RML, which then calls the errmgr for instructions.
* opal_buffer_t now conforms to standard object rules re OBJ_RETAIN as we no longer "unload" the incoming object
* NIC failure is reported to the TCP component, which then tries to resend the message across any other available TCP NIC. If that doesn't work, then the message is given back to the OOB base to try using other components. If all that fails, then the error is reported to the RML, which reports to the errmgr for instructions
* obviously from the above, multiple OOB components (e.g., TCP and UD) can be active in parallel
* the matching code has been moved to the RML (and out of the OOB/TCP component) so it is independent of transport
* routing is done by the individual OOB modules (as opposed to the RML). Thus, both routed and non-routed transports can simultaneously be active
* all blocking send/recv APIs have been removed. Everything operates asynchronously.
KNOWN LIMITATIONS:
* although provision is made for component failover as described above, the code for doing so has not been fully implemented yet. At the moment, if all connections for a given peer fail, the errmgr is notified of a "lost connection", which by default results in termination of the job if it was a lifeline
* the IPv6 code is present and compiles, but is not complete. Since the current IPv6 support in the OOB doesn't work anyway, I don't consider this a blocker
* routing is performed at the individual module level, yet the active routed component is selected on a global basis. We probably should update that to reflect that different transports may need/choose to route in different ways
* obviously, not every error path has been tested nor necessarily covered
* determining abnormal termination is more challenging than in the old code as we now potentially have multiple ways of connecting to a process. Ideally, we would declare "connection failed" when *all* transports can no longer reach the process, but that requires some additional (possibly complex) code. For now, the code replicates the old behavior only somewhat modified - i.e., if a module sees its connection fail, it checks to see if it is a lifeline. If so, it notifies the errmgr that the lifeline is lost - otherwise, it notifies the errmgr that a non-lifeline connection was lost.
* reachability is determined solely on the basis of a shared subnet address/mask - more sophisticated algorithms (e.g., the one used in the tcp btl) are required to handle routing via gateways
* the RML needs to assign sequence numbers to each message on a per-peer basis. The receiving RML will then deliver messages in order, thus preventing out-of-order messaging in the case where messages travel across different transports or a message needs to be redirected/resent due to failure of a NIC
This commit was SVN r29058.
2013-08-22 16:37:40 +00:00
|
|
|
#define ompi_rte_send_buffer_nb(a, b, c, d, e) orte_rml.send_buffer_nb(a, b, c, d, e)
|
2013-01-27 23:25:10 +00:00
|
|
|
#define ompi_rte_recv_buffer_nb(a, b, c, d, e) orte_rml.recv_buffer_nb(a, b, c, d, e)
|
|
|
|
|
#define ompi_rte_recv_cancel(a, b) orte_rml.recv_cancel(a, b)
|
|
|
|
|
#define ompi_rte_parse_uris(a, b, c) orte_rml_base_parse_uris(a, b, c)
|
As per the RFC, bring in the ORTE async progress code and the rewrite of OOB:
*** THIS RFC INCLUDES A MINOR CHANGE TO THE MPI-RTE INTERFACE ***
Note: during the course of this work, it was necessary to completely separate the MPI and RTE progress engines. There were multiple places in the MPI layer where ORTE_WAIT_FOR_COMPLETION was being used. A new OMPI_WAIT_FOR_COMPLETION macro was created (defined in ompi/mca/rte/rte.h) that simply cycles across opal_progress until the provided flag becomes false. Places where the MPI layer blocked waiting for RTE to complete an event have been modified to use this macro.
***************************************************************************************
I am reissuing this RFC because of the time that has passed since its original release. Since its initial release and review, I have debugged it further to ensure it fully supports tests like loop_spawn. It therefore seems ready for merge back to the trunk. Given its prior review, I have set the timeout for one week.
The code is in https://bitbucket.org/rhc/ompi-oob2
WHAT: Rewrite of ORTE OOB
WHY: Support asynchronous progress and a host of other features
WHEN: Wed, August 21
SYNOPSIS:
The current OOB has served us well, but a number of limitations have been identified over the years. Specifically:
* it is only progressed when called via opal_progress, which can lead to hangs or recursive calls into libevent (which is not supported by that code)
* we've had issues when multiple NICs are available as the code doesn't "shift" messages between transports - thus, all nodes had to be available via the same TCP interface.
* the OOB "unloads" incoming opal_buffer_t objects during the transmission, thus preventing use of OBJ_RETAIN in the code when repeatedly sending the same message to multiple recipients
* there is no failover mechanism across NICs - if the selected NIC (or its attached switch) fails, we are forced to abort
* only one transport (i.e., component) can be "active"
The revised OOB resolves these problems:
* async progress is used for all application processes, with the progress thread blocking in the event library
* each available TCP NIC is supported by its own TCP module. The ability to asynchronously progress each module independently is provided, but not enabled by default (a runtime MCA parameter turns it "on")
* multi-address TCP NICs (e.g., a NIC with both an IPv4 and IPv6 address, or with virtual interfaces) are supported - reachability is determined by comparing the contact info for a peer against all addresses within the range covered by the address/mask pairs for the NIC.
* a message that arrives on one TCP NIC is automatically shifted to whatever NIC that is connected to the next "hop" if that peer cannot be reached by the incoming NIC. If no TCP module will reach the peer, then the OOB attempts to send the message via all other available components - if none can reach the peer, then an "error" is reported back to the RML, which then calls the errmgr for instructions.
* opal_buffer_t now conforms to standard object rules re OBJ_RETAIN as we no longer "unload" the incoming object
* NIC failure is reported to the TCP component, which then tries to resend the message across any other available TCP NIC. If that doesn't work, then the message is given back to the OOB base to try using other components. If all that fails, then the error is reported to the RML, which reports to the errmgr for instructions
* obviously from the above, multiple OOB components (e.g., TCP and UD) can be active in parallel
* the matching code has been moved to the RML (and out of the OOB/TCP component) so it is independent of transport
* routing is done by the individual OOB modules (as opposed to the RML). Thus, both routed and non-routed transports can simultaneously be active
* all blocking send/recv APIs have been removed. Everything operates asynchronously.
KNOWN LIMITATIONS:
* although provision is made for component failover as described above, the code for doing so has not been fully implemented yet. At the moment, if all connections for a given peer fail, the errmgr is notified of a "lost connection", which by default results in termination of the job if it was a lifeline
* the IPv6 code is present and compiles, but is not complete. Since the current IPv6 support in the OOB doesn't work anyway, I don't consider this a blocker
* routing is performed at the individual module level, yet the active routed component is selected on a global basis. We probably should update that to reflect that different transports may need/choose to route in different ways
* obviously, not every error path has been tested nor necessarily covered
* determining abnormal termination is more challenging than in the old code as we now potentially have multiple ways of connecting to a process. Ideally, we would declare "connection failed" when *all* transports can no longer reach the process, but that requires some additional (possibly complex) code. For now, the code replicates the old behavior only somewhat modified - i.e., if a module sees its connection fail, it checks to see if it is a lifeline. If so, it notifies the errmgr that the lifeline is lost - otherwise, it notifies the errmgr that a non-lifeline connection was lost.
* reachability is determined solely on the basis of a shared subnet address/mask - more sophisticated algorithms (e.g., the one used in the tcp btl) are required to handle routing via gateways
* the RML needs to assign sequence numbers to each message on a per-peer basis. The receiving RML will then deliver messages in order, thus preventing out-of-order messaging in the case where messages travel across different transports or a message needs to be redirected/resent due to failure of a NIC
This commit was SVN r29058.
2013-08-22 16:37:40 +00:00
|
|
|
#define ompi_rte_send_cbfunc orte_rml_send_callback
|
2013-01-27 23:25:10 +00:00
|
|
|
|
|
|
|
|
/* Communication tags */
|
|
|
|
|
/* carry over the INVALID def */
|
|
|
|
|
#define OMPI_RML_TAG_INVALID ORTE_RML_TAG_INVALID
|
|
|
|
|
/* define a starting point to avoid conflicts */
|
|
|
|
|
#define OMPI_RML_TAG_BASE ORTE_RML_TAG_MAX
|
|
|
|
|
|
As per the RFC, bring in the ORTE async progress code and the rewrite of OOB:
*** THIS RFC INCLUDES A MINOR CHANGE TO THE MPI-RTE INTERFACE ***
Note: during the course of this work, it was necessary to completely separate the MPI and RTE progress engines. There were multiple places in the MPI layer where ORTE_WAIT_FOR_COMPLETION was being used. A new OMPI_WAIT_FOR_COMPLETION macro was created (defined in ompi/mca/rte/rte.h) that simply cycles across opal_progress until the provided flag becomes false. Places where the MPI layer blocked waiting for RTE to complete an event have been modified to use this macro.
***************************************************************************************
I am reissuing this RFC because of the time that has passed since its original release. Since its initial release and review, I have debugged it further to ensure it fully supports tests like loop_spawn. It therefore seems ready for merge back to the trunk. Given its prior review, I have set the timeout for one week.
The code is in https://bitbucket.org/rhc/ompi-oob2
WHAT: Rewrite of ORTE OOB
WHY: Support asynchronous progress and a host of other features
WHEN: Wed, August 21
SYNOPSIS:
The current OOB has served us well, but a number of limitations have been identified over the years. Specifically:
* it is only progressed when called via opal_progress, which can lead to hangs or recursive calls into libevent (which is not supported by that code)
* we've had issues when multiple NICs are available as the code doesn't "shift" messages between transports - thus, all nodes had to be available via the same TCP interface.
* the OOB "unloads" incoming opal_buffer_t objects during the transmission, thus preventing use of OBJ_RETAIN in the code when repeatedly sending the same message to multiple recipients
* there is no failover mechanism across NICs - if the selected NIC (or its attached switch) fails, we are forced to abort
* only one transport (i.e., component) can be "active"
The revised OOB resolves these problems:
* async progress is used for all application processes, with the progress thread blocking in the event library
* each available TCP NIC is supported by its own TCP module. The ability to asynchronously progress each module independently is provided, but not enabled by default (a runtime MCA parameter turns it "on")
* multi-address TCP NICs (e.g., a NIC with both an IPv4 and IPv6 address, or with virtual interfaces) are supported - reachability is determined by comparing the contact info for a peer against all addresses within the range covered by the address/mask pairs for the NIC.
* a message that arrives on one TCP NIC is automatically shifted to whatever NIC that is connected to the next "hop" if that peer cannot be reached by the incoming NIC. If no TCP module will reach the peer, then the OOB attempts to send the message via all other available components - if none can reach the peer, then an "error" is reported back to the RML, which then calls the errmgr for instructions.
* opal_buffer_t now conforms to standard object rules re OBJ_RETAIN as we no longer "unload" the incoming object
* NIC failure is reported to the TCP component, which then tries to resend the message across any other available TCP NIC. If that doesn't work, then the message is given back to the OOB base to try using other components. If all that fails, then the error is reported to the RML, which reports to the errmgr for instructions
* obviously from the above, multiple OOB components (e.g., TCP and UD) can be active in parallel
* the matching code has been moved to the RML (and out of the OOB/TCP component) so it is independent of transport
* routing is done by the individual OOB modules (as opposed to the RML). Thus, both routed and non-routed transports can simultaneously be active
* all blocking send/recv APIs have been removed. Everything operates asynchronously.
KNOWN LIMITATIONS:
* although provision is made for component failover as described above, the code for doing so has not been fully implemented yet. At the moment, if all connections for a given peer fail, the errmgr is notified of a "lost connection", which by default results in termination of the job if it was a lifeline
* the IPv6 code is present and compiles, but is not complete. Since the current IPv6 support in the OOB doesn't work anyway, I don't consider this a blocker
* routing is performed at the individual module level, yet the active routed component is selected on a global basis. We probably should update that to reflect that different transports may need/choose to route in different ways
* obviously, not every error path has been tested nor necessarily covered
* determining abnormal termination is more challenging than in the old code as we now potentially have multiple ways of connecting to a process. Ideally, we would declare "connection failed" when *all* transports can no longer reach the process, but that requires some additional (possibly complex) code. For now, the code replicates the old behavior only somewhat modified - i.e., if a module sees its connection fail, it checks to see if it is a lifeline. If so, it notifies the errmgr that the lifeline is lost - otherwise, it notifies the errmgr that a non-lifeline connection was lost.
* reachability is determined solely on the basis of a shared subnet address/mask - more sophisticated algorithms (e.g., the one used in the tcp btl) are required to handle routing via gateways
* the RML needs to assign sequence numbers to each message on a per-peer basis. The receiving RML will then deliver messages in order, thus preventing out-of-order messaging in the case where messages travel across different transports or a message needs to be redirected/resent due to failure of a NIC
This commit was SVN r29058.
2013-08-22 16:37:40 +00:00
|
|
|
#define OMPI_RML_PERSISTENT ORTE_RML_PERSISTENT
|
|
|
|
|
#define OMPI_RML_NON_PERSISTENT ORTE_RML_NON_PERSISTENT
|
2013-01-27 23:25:10 +00:00
|
|
|
|
For large scale systems, we would like to avoid doing a full modex during MPI_Init so that launch will scale a little better. At the moment, our options are somewhat limited as only a few BTLs don't immediately call modex_recv on all procs during startup. However, for those situations where someone can take advantage of it, add the ability to do a "modex on demand" retrieval of data from remote procs when we launch via mpirun.
NOTE: launch performance will be absolutely awful if you do this with BTLs that aren't configured to modex_recv on first message!
Even with "modex on demand", we still have to do a barrier in place of the modex - we simply don't move any data around, which does reduce the time impact. The barrier is required to ensure that the other proc has in fact registered all its BTL info and therefore is prepared to hand over a complete data package. Otherwise, you may not get the info you need. In addition, the shared memory BTL can fail to properly rendezvous as it expects the barrier to be in place.
This behavior will *only* take effect under the following conditions:
1. launched via mpirun
2. #procs is greater than ompi_hostname_cutoff, which defaults to UINT32_MAX
3. mca param rte_orte_direct_modex is set to 1. At the moment, we are having problems getting this param to register properly, so only the first two conditions are in effect. Still, the bottom line is you have to *want* this behavior to get it.
The planned next evolution of this will be to make the direct modex be non-blocking - this will require two fixes:
1. if the remote proc doesn't have the required info, then let it delay its response until it does. This means we need a way for the MPI layer to tell the RTE "I am done entering modex data".
2. adjust the SM rendezvous logic to loop until the required file has been created
Creating a placeholder to bring this over to 1.7.5 when ready.
cmr=v1.7.5:reviewer=hjelmn:subject=Enable direct modex at scale
This commit was SVN r30259.
2014-01-11 17:36:06 +00:00
|
|
|
/* define a local variable shared between component and module */
|
|
|
|
|
OMPI_MODULE_DECLSPEC extern bool ompi_rte_orte_direct_modex;
|
|
|
|
|
|
2013-01-27 23:25:10 +00:00
|
|
|
END_C_DECLS
|
|
|
|
|
|
|
|
|
|
#endif /* MCA_OMPI_RTE_ORTE_H */
|