1
1
Ralph Castain aec5cd08bd Per the PMIx RFC:
WHAT:    Merge the PMIx branch into the devel repo, creating a new
               OPAL “lmix” framework to abstract PMI support for all RTEs.
               Replace the ORTE daemon-level collectives with a new PMIx
               server and update the ORTE grpcomm framework to support
               server-to-server collectives

WHY:      We’ve had problems dealing with variations in PMI implementations,
               and need to extend the existing PMI definitions to meet exascale
               requirements.

WHEN:   Mon, Aug 25

WHERE:  https://github.com/rhc54/ompi-svn-mirror.git

Several community members have been working on a refactoring of the current PMI support within OMPI. Although the APIs are common, Slurm and Cray implement a different range of capabilities, and package them differently. For example, Cray provides an integrated PMI-1/2 library, while Slurm separates the two and requires the user to specify the one to be used at runtime. In addition, several bugs in the Slurm implementations have caused problems requiring extra coding.

All this has led to a slew of #if’s in the PMI code and bugs when the corner-case logic for one implementation accidentally traps the other. Extending this support to other implementations would have increased this complexity to an unacceptable level.

Accordingly, we have:

* created a new OPAL “pmix” framework to abstract the PMI support, with separate components for Cray, Slurm PMI-1, and Slurm PMI-2 implementations.

* Replaced the current ORTE grpcomm daemon-based collective operation with an integrated PMIx server, and updated the grpcomm APIs to provide more flexible, multi-algorithm support for collective operations. At this time, only the xcast and allgather operations are supported.

* Replaced the current global collective id with a signature based on the names of the participating procs. The allows an unlimited number of collectives to be executed by any group of processes, subject to the requirement that only one collective can be active at a time for a unique combination of procs. Note that a proc can be involved in any number of simultaneous collectives - it is the specific combination of procs that is subject to the constraint

* removed the prior OMPI/OPAL modex code

* added new macros for executing modex send/recv to simplify use of the new APIs. The send macros allow the caller to specify whether or not the BTL supports async modex operations - if so, then the non-blocking “fence” operation is used, if the active PMIx component supports it. Otherwise, the default is a full blocking modex exchange as we currently perform.

* retained the current flag that directs us to use a blocking fence operation, but only to retrieve data upon demand

This commit was SVN r32570.
2014-08-21 18:56:47 +00:00

107 строки
3.4 KiB
C

/*
* Copyright (c) 2010 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2012-2013 Los Alamos National Security, Inc. All rights reserved.
* Copyright (c) 2013-2014 Intel, Inc. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
/** @file:
*/
#ifndef MCA_DSTORE_BASE_H
#define MCA_DSTORE_BASE_H
#include "opal_config.h"
#include "opal/types.h"
#include "opal/mca/mca.h"
#include "opal/mca/base/mca_base_framework.h"
#include "opal/mca/event/event.h"
#include "opal/class/opal_hash_table.h"
#include "opal/class/opal_list.h"
#include "opal/class/opal_pointer_array.h"
#include "opal/dss/dss.h"
#include "opal/mca/dstore/dstore.h"
BEGIN_C_DECLS
OPAL_DECLSPEC extern mca_base_framework_t opal_dstore_base_framework;
/**
* Select a dstore module
*/
OPAL_DECLSPEC int opal_dstore_base_select(void);
/* DSTORE is an oddball framework in that it:
*
* has an active storage component that issues handle-specific
* modules. This is done to provide separate storage areas that
* are isolated from each other, and thus don't have to worry
* about overlapping keys
*
* a backfill module used to attempt to retrieve data that has
* been requested, but that the handle-specific storage module
* does not contain. This is used in situations where data has
* not been provided at startup, and we need to retrieve it
* solely on-demand
*/
typedef struct {
opal_dstore_base_component_t *storage_component;
opal_dstore_base_module_t *backfill_module;
opal_pointer_array_t handles; // array of open datastore handles
} opal_dstore_base_t;
OPAL_DECLSPEC extern opal_dstore_base_t opal_dstore_base;
typedef struct {
opal_object_t super;
char *name;
opal_dstore_base_module_t *module;
} opal_dstore_handle_t;
OBJ_CLASS_DECLARATION(opal_dstore_handle_t);
/**
* Data for a particular opal process
* The name association is maintained in the
* proc_data hash table.
*/
typedef struct {
/** Structure can be put on lists (including in hash tables) */
opal_list_item_t super;
bool loaded;
/* List of opal_value_t structures containing all data
received from this process, sorted by key. */
opal_list_t data;
} opal_dstore_proc_data_t;
OBJ_CLASS_DECLARATION(opal_dstore_proc_data_t);
OPAL_DECLSPEC int opal_dstore_base_open(const char *name);
OPAL_DECLSPEC int opal_dstore_base_close(int dstorehandle);
OPAL_DECLSPEC int opal_dstore_base_store(int dstorehandle,
const opal_identifier_t *id,
opal_value_t *kv);
OPAL_DECLSPEC int opal_dstore_base_fetch(int dstorehandle,
const opal_identifier_t *id,
const char *key,
opal_list_t *kvs);
OPAL_DECLSPEC int opal_dstore_base_remove_data(int dstorehandle,
const opal_identifier_t *id,
const char *key);
/* support */
OPAL_DECLSPEC opal_dstore_proc_data_t* opal_dstore_base_lookup_proc(opal_hash_table_t *jtable,
opal_identifier_t id);
OPAL_DECLSPEC opal_value_t* opal_dstore_base_lookup_keyval(opal_dstore_proc_data_t *proc_data,
const char *key);
END_C_DECLS
#endif