1
1
openmpi/orte/runtime/runtime.h

160 строки
5.3 KiB
C
Исходник Обычный вид История

/*
* Copyright (c) 2004-2005 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2006 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2007 Sun Microsystems, Inc. All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
/**
* @file
*
* Interface into the Open MPI Run Time Environment
*/
#ifndef ORTE_RUNTIME_H
#define ORTE_RUNTIME_H
#include "orte_config.h"
#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif
#include "orte/mca/gpr/gpr_types.h"
#include "opal/util/cmd_line.h"
2004-09-23 14:40:05 +00:00
#include "orte/util/univ_info.h"
#include "orte/mca/ns/ns.h"
Commit the orted-failed-to-start code. This correctly causes the system to detect the failure of an orted to start and allows the system to terminate all procs/orteds that *did* start. The primary change that underlies all this is in the OOB. Specifically, the problem in the code until now has been that the OOB attempts to resolve an address when we call the "send" to an unknown recipient. The OOB would then wait forever if that recipient never actually started (and hence, never reported back its OOB contact info). In the case of an orted that failed to start, we would correctly detect that the orted hadn't started, but then we would attempt to order all orteds (including the one that failed to start) to die. This would cause the OOB to "hang" the system. Unfortunately, revising how the OOB resolves addresses introduced a number of additional problems. Specifically, and most troublesome, was the fact that comm_spawn involved the immediate transmission of the rendezvous point from parent-to-child after the child was spawned. The current code used the OOB address resolution as a "barrier" - basically, the parent would attempt to send the info to the child, and then "hold" there until the child's contact info had arrived (meaning the child had started) and the send could be completed. Note that this also caused comm_spawn to "hang" the entire system if the child never started... The app-failed-to-start helped improve that behavior - this code provides additional relief. With this change, the OOB will return an ADDRESSEE_UNKNOWN error if you attempt to send to a recipient whose contact info isn't already in the OOB's hash tables. To resolve comm_spawn issues, we also now force the cross-sharing of connection info between parent and child jobs during spawn. Finally, to aid in setting triggers to the right values, we introduce the "arith" API for the GPR. This function allows you to atomically change the value in a registry location (either divide, multiply, add, or subtract) by the provided operand. It is equivalent to first fetching the value using a "get", then modifying it, and then putting the result back into the registry via a "put". This commit was SVN r14711.
2007-05-21 18:31:28 +00:00
/* some convenience definitions for code clarity */
#define ORTE_INFRASTRUCTURE true
#define ORTE_NON_INFRASTRUCTURE false
#if defined(c_plusplus) || defined(__cplusplus)
extern "C" {
#endif
/**
* Abort the current application
*
* Aborts currently running application, NOTE: We do NOT call the
* regular C-library "abort" function, even
* though that would have alerted us to the fact that this is
* an abnormal termination, because it would automatically cause
* a core file to be generated. The "report" flag indicates if the
* function should create an appropriate file to alert the local
* orted that termination was abnormal.
*/
ORTE_DECLSPEC void orte_abort(int status, bool report) __opal_attribute_noreturn__;
/**
* Initialize the Open Run Time Environment
*
* Initlize the Open Run Time Environment, including process
* control, malloc debugging and threads, and out of band messaging.
* This function should be called exactly once. This function should
* be called by every application using the RTE interface, including
* MPI applications and mpirun.
*
* @param infrastructure Whether we are ORTE infrastructure or an ORTE
* application
*/
These changes were mostly captured in a prior RFC (except for #2 below) and are aimed specifically at improving startup performance and setting up the remaining modifications described in that RFC. The commit has been tested for C/R and Cray operations, and on Odin (SLURM, rsh) and RoadRunner (TM). I tried to update all environments, but obviously could not test them. I know that Windows needs some work, and have highlighted what is know to be needed in the odls process component. This represents a lot of work by Brian, Tim P, Josh, and myself, with much advice from Jeff and others. For posterity, I have appended a copy of the email describing the work that was done: As we have repeatedly noted, the modex operation in MPI_Init is the single greatest consumer of time during startup. To-date, we have executed that operation as an ORTE stage gate that held the process until a startup message containing all required modex (and OOB contact info - see #3 below) info could be sent to it. Each process would send its data to the HNP's registry, which assembled and sent the message when all processes had reported in. In addition, ORTE had taken responsibility for monitoring process status as it progressed through a series of "stage gates". The process reported its status at each gate, and ORTE would then send a "release" message once all procs had reported in. The incoming changes revamp these procedures in three ways: 1. eliminating the ORTE stage gate system and cleanly delineating responsibility between the OMPI and ORTE layers for MPI init/finalize. The modex stage gate (STG1) has been replaced by a collective operation in the modex itself that performs an allgather on the required modex info. The allgather is implemented using the orte_grpcomm framework since the BTL's are not active at that point. At the moment, the grpcomm framework only has a "basic" component analogous to OMPI's "basic" coll framework - I would recommend that the MPI team create additional, more advanced components to improve performance of this step. The other stage gates have been replaced by orte_grpcomm barrier functions. We tried to use MPI barriers instead (since the BTL's are active at that point), but - as we discussed on the telecon - these are not currently true barriers so the job would hang when we fell through while messages were still in process. Note that the grpcomm barrier doesn't actually resolve that problem, but Brian has pointed out that we are unlikely to ever see it violated. Again, you might want to spend a little time on an advanced barrier algorithm as the one in "basic" is very simplistic. Summarizing this change: ORTE no longer tracks process state nor has direct responsibility for synchronizing jobs. This is now done via collective operations within the MPI layer, albeit using ORTE collective communication services. I -strongly- urge the MPI team to implement advanced collective algorithms to improve the performance of this critical procedure. 2. reducing the volume of data exchanged during modex. Data in the modex consisted of the process name, the name of the node where that process is located (expressed as a string), plus a string representation of all contact info. The nodename was required in order for the modex to determine if the process was local or not - in addition, some people like to have it to print pretty error messages when a connection failed. The size of this data has been reduced in three ways: (a) reducing the size of the process name itself. The process name consisted of two 32-bit fields for the jobid and vpid. This is far larger than any current system, or system likely to exist in the near future, can support. Accordingly, the default size of these fields has been reduced to 16-bits, which means you can have 32k procs in each of 32k jobs. Since the daemons must have a vpid, and we require one daemon/node, this also restricts the default configuration to 32k nodes. To support any future "mega-clusters", a configuration option --enable-jumbo-apps has been added. This option increases the jobid and vpid field sizes to 32-bits. Someday, if necessary, someone can add yet another option to increase them to 64-bits, I suppose. (b) replacing the string nodename with an integer nodeid. Since we have one daemon/node, the nodeid corresponds to the local daemon's vpid. This replaces an often lengthy string with only 2 (or at most 4) bytes, a substantial reduction. (c) when the mca param requesting that nodenames be sent to support pretty error messages, a second mca param is now used to request FQDN - otherwise, the domain name is stripped (by default) from the message to save space. If someone wants to combine those into a single param somehow (perhaps with an argument?), they are welcome to do so - I didn't want to alter what people are already using. While these may seem like small savings, they actually amount to a significant impact when aggregated across the entire modex operation. Since every proc must receive the modex data regardless of the collective used to send it, just reducing the size of the process name removes nearly 400MBytes of communication from a 32k proc job (admittedly, much of this comm may occur in parallel). So it does add up pretty quickly. 3. routing RML messages to reduce connections. The default messaging system remains point-to-point - i.e., each proc opens a socket to every proc it communicates with and sends its messages directly. A new option uses the orteds as routers - i.e., each proc only opens a single socket to its local orted. All messages are sent from the proc to the orted, which forwards the message to the orted on the node where the intended recipient proc is located - that orted then forwards the message to its local proc (the recipient). This greatly reduces the connection storm we have encountered during startup. It also has the benefit of removing the sharing of every proc's OOB contact with every other proc. The orted routing tables are populated during launch since every orted gets a map of where every proc is being placed. Each proc, therefore, only needs to know the contact info for its local daemon, which is passed in via the environment when the proc is fork/exec'd by the daemon. This alone removes ~50 bytes/process of communication that was in the current STG1 startup message - so for our 32k proc job, this saves us roughly 32k*50 = 1.6MBytes sent to 32k procs = 51GBytes of messaging. Note that you can use the new routing method by specifying -mca routed tree - if you so desire. This mode will become the default at some point in the future. There are a few minor additional changes in the commit that I'll just note in passing: * propagation of command line mca params to the orteds - fixes ticket #1073. See note there for details. * requiring of "finalize" prior to "exit" for MPI procs - fixes ticket #1144. See note there for details. * cleanup of some stale header files This commit was SVN r16364.
2007-10-05 19:48:23 +00:00
ORTE_DECLSPEC int orte_init(bool infrastructure);
/**
* Initialize parameters for ORTE.
*
* @retval ORTE_SUCCESS Upon success.
* @retval ORTE_ERROR Upon failure.
*/
ORTE_DECLSPEC int orte_register_params(bool infrastructure);
/**
* Re-init the Open run time environment.
*
* Restart selected components with a new process name.
*/
ORTE_DECLSPEC int orte_restart(orte_process_name_t* name, const char* uri);
/**
* Finalize the Open run time environment. Any function calling \code
* orte_init should call \code orte_finalize.
*
2004-09-23 14:40:05 +00:00
*/
ORTE_DECLSPEC int orte_finalize(void);
/*
* Change state as processes complete registration/unregistration
*/
ORTE_DECLSPEC void orte_all_procs_registered(orte_gpr_notify_message_t* match, void* cbdata);
ORTE_DECLSPEC void orte_all_procs_unregistered(orte_gpr_notify_message_t* match, void* cbdata);
ORTE_DECLSPEC int orte_monitor_procs_registered(void);
ORTE_DECLSPEC int orte_monitor_procs_unregistered(void);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-28 21:03:31 +00:00
/**
* Obtain a listing of all the universes on the machine
*
* @param univ_list An opal_list_t is returned to the user.
* This is not initalized in the function, the caller retains
* the responsibility for this variable.
* @retval ORTE_SUCCESS Upon successful search.
* @retval ORTE_ERROR Upon unsuccessful search.
*/
ORTE_DECLSPEC int orte_universe_search(opal_list_t *universe_list, bool report_broken_files,
bool remove_broken_files);
Brining over the session directory and universe changes from the tmp/jjhursey-ft-cr branch. In this commit we change the way universe names are created. Before we by default first created "default-universe" then if there was a conflict we created "default-universe-PID" where PID is the PID of the HNP. Now we create "default-universe-PID" all the time (when a default universe name is used). This makes it much easier when trying to find a HNP from an outside app (e.g. orte-ps, orteconsole, ...) This also adds a "search" function to find all of the universes on the machine. This is useful in many contexts when trying to find a persistent daemon or when trying to connect to a HNP. This commit also makes orte_universe_t an opal_object_t, which is something that needed to happen, and only effected the SDS in one of it's base functions. I was asked to bring this over to aid in fixing orteconsole and orteprobe. Due to the change of orte_universe_t to an object orteprobe may need to be updated to reflect this change. Since orteprobe needs to be looked at anyway I'll leave this to Ralph to take care of. *Note*: These changes do not depend upon any of the FT work (but the FT work does depend upon them). These were brought over to help in fixing some of the ORTE tool set that require the functionality layed out in this patch. Testing: Ran the 'ibm' tests before and after this change, and all was as well as before the change. If anyone notices additional irregularities in the system let me know. But none are expected. This commit was SVN r10550.
2006-06-28 21:03:31 +00:00
/**
* Check for universe existence
*
Some of these didn't really change - I was just in/out of them for diagnostics while chasing a bug. Got caught by my good buddy Tim again :) on his parse_contact_info function, which requires that the space for the answer be allocated in advance. Sigh. Anyway, mpirun2 now works again. My apologies if you tried it in the last few hours and found it didn't. Also removed the mpirun3 directory since we are basically dragging mpirun2 along with us - no need to create a new version after all. Made a few changes to the universe info structure, eliminating the "webserver" and "socket" fields since we will do those contacts through the oob channel. Also changed the "silent_mode" field to "console" since silent mode is the default - the flag needs to tell you to turn the console on, not off. Parse environ function now gets the ns and gpr replica contact info and loads it in the proper places to hand it off to the respective components, thus allowing me to check connection to them as part of determining if the named universe already exists. Changed the local_universe_exists function accordingly and gave it a new name (since the replicas may not be local). This name will shortly be changed to "ompi_rte_join_universe" as I complete the logic for doing that function. Please let me know if you see any problems. I successfully ran some trivial multi-process functions in both mpirun2 and singleton modes, and ran the seed daemon as well, so I think it should all be okay. This commit was SVN r2611.
2004-09-11 12:56:52 +00:00
* Checks to see if a specified universe exists. If so, attempts
* to connect to verify that the universe is accepting connections.
Some of these didn't really change - I was just in/out of them for diagnostics while chasing a bug. Got caught by my good buddy Tim again :) on his parse_contact_info function, which requires that the space for the answer be allocated in advance. Sigh. Anyway, mpirun2 now works again. My apologies if you tried it in the last few hours and found it didn't. Also removed the mpirun3 directory since we are basically dragging mpirun2 along with us - no need to create a new version after all. Made a few changes to the universe info structure, eliminating the "webserver" and "socket" fields since we will do those contacts through the oob channel. Also changed the "silent_mode" field to "console" since silent mode is the default - the flag needs to tell you to turn the console on, not off. Parse environ function now gets the ns and gpr replica contact info and loads it in the proper places to hand it off to the respective components, thus allowing me to check connection to them as part of determining if the named universe already exists. Changed the local_universe_exists function accordingly and gave it a new name (since the replicas may not be local). This name will shortly be changed to "ompi_rte_join_universe" as I complete the logic for doing that function. Please let me know if you see any problems. I successfully ran some trivial multi-process functions in both mpirun2 and singleton modes, and ran the seed daemon as well, so I think it should all be okay. This commit was SVN r2611.
2004-09-11 12:56:52 +00:00
* If both ns and gpr replicas provided, first checks for those
* connections. Gets any missing info from the universe contact.
*
* @param univ Pointer to universe info struct where any found info
* is to be stored
*
* @retval ORTE_SUCCESS Universe found and connection accepted
Commit the orted-failed-to-start code. This correctly causes the system to detect the failure of an orted to start and allows the system to terminate all procs/orteds that *did* start. The primary change that underlies all this is in the OOB. Specifically, the problem in the code until now has been that the OOB attempts to resolve an address when we call the "send" to an unknown recipient. The OOB would then wait forever if that recipient never actually started (and hence, never reported back its OOB contact info). In the case of an orted that failed to start, we would correctly detect that the orted hadn't started, but then we would attempt to order all orteds (including the one that failed to start) to die. This would cause the OOB to "hang" the system. Unfortunately, revising how the OOB resolves addresses introduced a number of additional problems. Specifically, and most troublesome, was the fact that comm_spawn involved the immediate transmission of the rendezvous point from parent-to-child after the child was spawned. The current code used the OOB address resolution as a "barrier" - basically, the parent would attempt to send the info to the child, and then "hold" there until the child's contact info had arrived (meaning the child had started) and the send could be completed. Note that this also caused comm_spawn to "hang" the entire system if the child never started... The app-failed-to-start helped improve that behavior - this code provides additional relief. With this change, the OOB will return an ADDRESSEE_UNKNOWN error if you attempt to send to a recipient whose contact info isn't already in the OOB's hash tables. To resolve comm_spawn issues, we also now force the cross-sharing of connection info between parent and child jobs during spawn. Finally, to aid in setting triggers to the right values, we introduce the "arith" API for the GPR. This function allows you to atomically change the value in a registry location (either divide, multiply, add, or subtract) by the provided operand. It is equivalent to first fetching the value using a "get", then modifying it, and then putting the result back into the registry via a "put". This commit was SVN r14711.
2007-05-21 18:31:28 +00:00
* @retval ORTE_ERR_NO_CONNECTION_ALLOWED Universe found, but not persistent or
* restricted to local scope
Commit the orted-failed-to-start code. This correctly causes the system to detect the failure of an orted to start and allows the system to terminate all procs/orteds that *did* start. The primary change that underlies all this is in the OOB. Specifically, the problem in the code until now has been that the OOB attempts to resolve an address when we call the "send" to an unknown recipient. The OOB would then wait forever if that recipient never actually started (and hence, never reported back its OOB contact info). In the case of an orted that failed to start, we would correctly detect that the orted hadn't started, but then we would attempt to order all orteds (including the one that failed to start) to die. This would cause the OOB to "hang" the system. Unfortunately, revising how the OOB resolves addresses introduced a number of additional problems. Specifically, and most troublesome, was the fact that comm_spawn involved the immediate transmission of the rendezvous point from parent-to-child after the child was spawned. The current code used the OOB address resolution as a "barrier" - basically, the parent would attempt to send the info to the child, and then "hold" there until the child's contact info had arrived (meaning the child had started) and the send could be completed. Note that this also caused comm_spawn to "hang" the entire system if the child never started... The app-failed-to-start helped improve that behavior - this code provides additional relief. With this change, the OOB will return an ADDRESSEE_UNKNOWN error if you attempt to send to a recipient whose contact info isn't already in the OOB's hash tables. To resolve comm_spawn issues, we also now force the cross-sharing of connection info between parent and child jobs during spawn. Finally, to aid in setting triggers to the right values, we introduce the "arith" API for the GPR. This function allows you to atomically change the value in a registry location (either divide, multiply, add, or subtract) by the provided operand. It is equivalent to first fetching the value using a "get", then modifying it, and then putting the result back into the registry via a "put". This commit was SVN r14711.
2007-05-21 18:31:28 +00:00
* @retval ORTE_ERR_CONNECTION_FAILED Universe found, but connection attempt
* failed. Probably caused by unclean termination of the universe seed
* daemon.
*/
ORTE_DECLSPEC int orte_universe_exists(orte_universe_t *univ);
/**
* Establish a Head Node Process on a cluster's front end
*/
ORTE_DECLSPEC int orte_setup_hnp(char *target_cluster, char *headnode, char *username);
/**
* Clean out all directories in a session directory except for the one
* handed in.
* @param my_universe Name of universe to not remove
* @param verbose Print out information as directories are removed
*/
ORTE_DECLSPEC void orte_universe_clean_directories(char *my_universe, int verbose);
#if defined(c_plusplus) || defined(__cplusplus)
}
#endif
Commit the orted-failed-to-start code. This correctly causes the system to detect the failure of an orted to start and allows the system to terminate all procs/orteds that *did* start. The primary change that underlies all this is in the OOB. Specifically, the problem in the code until now has been that the OOB attempts to resolve an address when we call the "send" to an unknown recipient. The OOB would then wait forever if that recipient never actually started (and hence, never reported back its OOB contact info). In the case of an orted that failed to start, we would correctly detect that the orted hadn't started, but then we would attempt to order all orteds (including the one that failed to start) to die. This would cause the OOB to "hang" the system. Unfortunately, revising how the OOB resolves addresses introduced a number of additional problems. Specifically, and most troublesome, was the fact that comm_spawn involved the immediate transmission of the rendezvous point from parent-to-child after the child was spawned. The current code used the OOB address resolution as a "barrier" - basically, the parent would attempt to send the info to the child, and then "hold" there until the child's contact info had arrived (meaning the child had started) and the send could be completed. Note that this also caused comm_spawn to "hang" the entire system if the child never started... The app-failed-to-start helped improve that behavior - this code provides additional relief. With this change, the OOB will return an ADDRESSEE_UNKNOWN error if you attempt to send to a recipient whose contact info isn't already in the OOB's hash tables. To resolve comm_spawn issues, we also now force the cross-sharing of connection info between parent and child jobs during spawn. Finally, to aid in setting triggers to the right values, we introduce the "arith" API for the GPR. This function allows you to atomically change the value in a registry location (either divide, multiply, add, or subtract) by the provided operand. It is equivalent to first fetching the value using a "get", then modifying it, and then putting the result back into the registry via a "put". This commit was SVN r14711.
2007-05-21 18:31:28 +00:00
#endif /* RUNTIME_H */