#2183: LDMS: add stream for phase data #2184
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Pipelines resultsPR tests (gcc-12, ubuntu, mpich) Build for 1c98fba (2023-11-09 17:49:25 UTC) vt-build-amd64-ubuntu-20-04-clang-9-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-alpine-3-16-clang-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-24-04-gcc-14-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-10-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-10-openmpi-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-clang-15-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-gcc-12-vtk-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-clang-12-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-10-openmpi-cpp-spack Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-clang-11-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-24-04-clang-16-zoltan-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-gcc-12-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-gcc-11-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-icpx-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-24-04-clang-18-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-9-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-24-04-gcc-13-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-9-cuda-11-4-3-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-24-04-clang-17-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-24-04-clang-16-vtk-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-clang-10-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-9-cuda-12-2-0-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-clang-14-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-9-cpp-docs Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-22-04-clang-13-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) vt-build-amd64-ubuntu-20-04-gcc-9-ldms-cpp Build for cee9682 (2025-08-12 15:20:44 UTC) |
JacobDomagala
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
365191b to
cacd48a
Compare
JacobDomagala
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
2 times, most recently
from
8baa8fc to
7b49e32
Compare
Member
|
Code used for reading the data: #include <ldms/ldms.h>
#include <ldms/ldmsd_stream.h>
#include <ovis_util/util.h>
int stream_handler(ldmsd_stream_client_t c, void *ctxt,
ldmsd_stream_type_t stream_type,
const char *data, size_t data_len,
json_entity_t entity)
{
// Process the received data
printf("%.*s\n", (int)data_len, data);
return 0;
}
int main()
{
ldms_init(256);
ldms_t ldms = ldms_xprt_new_with_auth("sock", "none", NULL);
int rc = ldms_xprt_connect_by_name(ldms, "localhost", "10444", NULL, NULL);
if(!rc){
printf("Error code %d \n", rc);
}
ldmsd_stream_subscribe("LB_data", stream_handler, NULL);
while (1){
// Keep running to continue receiving data
}
ldms_xprt_put(ldms);
return 0;
}Config file: Script to launch the LDMS deamon: #!/bin/bash
TOP=/ovis/LDMS_install
export LD_LIBRARY_PATH=$TOP/lib/:$TOP/lib:$LD_LIBRARY_PATH
export LDMSD_PLUGIN_LIBPATH=$TOP/lib/ovis-ldms
export ZAP_LIBPATH=$TOP/lib/ovis-ldms
export PATH=$TOP/sbin:$TOP/bin:$PATH
export PYTHONPATH=$TOP/lib/python2.7/site-packages
ldmsd -x sock:10444 -c /ldmsd.conf -l /tmp/demo_ldmsd_log -v DEBUG -r $(pwd)/ldmsd.pid
|
JacobDomagala
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
5 times, most recently
from
caca0d1 to
1939ddb
Compare
lifflander
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
1939ddb to
1c98fba
Compare
Logan590
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
fb560c0 to
78c2a34
Compare
|
diff --git a/src/vt/configs/features/features_defines.h b/src/vt/configs/features/features_defines.h
index 3e645c2fb..39c01bbc1 100644
--- a/src/vt/configs/features/features_defines.h
+++ b/src/vt/configs/features/features_defines.h
@@ -63,7 +63,7 @@
#define vt_feature_memory_pool 0 || vt_feature_cmake_memory_pool
#define vt_feature_priorities 0 || vt_feature_cmake_priorities
#define vt_feature_fcontext 0 || vt_feature_cmake_fcontext
-#define vt_feature_ldms 0 || vt_feature_cmake_ldms
+#define vt_feature_ldms 0 || vt_feature_cmake_ldms
#define vt_feature_mimalloc 0 || vt_feature_cmake_mimalloc
#define vt_feature_mpi_access_guards 0 || vt_feature_cmake_mpi_access_guards
#define vt_feature_zoltan 0 || vt_feature_cmake_zoltan
diff --git a/src/vt/vrt/collection/balance/lb_invoke/lb_manager.cc b/src/vt/vrt/collection/balance/lb_invoke/lb_manager.cc
index 1f7f8adc8..0e23ab61b 100644
--- a/src/vt/vrt/collection/balance/lb_invoke/lb_manager.cc
+++ b/src/vt/vrt/collection/balance/lb_invoke/lb_manager.cc
@@ -631,11 +631,10 @@ void LBManager::stagePreLBStatistics(const StatisticMapType &statistics) {
nlohmann::json j;
j["pre-LB"] = lb::jsonifyPhaseStatistics(statistics);
-
- #if vt_check_enabled(ldms)
+#if vt_check_enabled(ldms)
j["ts"] = MPI_Wtime();
theNodeLBData()->writeJSONToLDMS(j);
- #endif
+#endif
if (!statistics_writer_) {
createStatisticsFile();
diff --git a/src/vt/vrt/collection/balance/node_lb_data.cc b/src/vt/vrt/collection/balance/node_lb_data.cc
index 2d87ed386..bff24dd3e 100644
--- a/src/vt/vrt/collection/balance/node_lb_data.cc
+++ b/src/vt/vrt/collection/balance/node_lb_data.cc
@@ -67,7 +67,6 @@
#include INCLUDE_FMT_FORMAT
-
namespace vt { namespace vrt { namespace collection { namespace balance {
void NodeLBData::setProxy(objgroup::proxy::Proxy<NodeLBData> in_proxy) {
@@ -169,8 +168,12 @@ void NodeLBData::initializeLDMS() {
const auto hostname = getenv("VT_LDMS_HOSTNAME");
const auto port = getenv("VT_LDMS_PORT");
- const auto returnCode = ldms_xprt_connect_by_name(ldms_, hostname, port, NULL, NULL);
- vtWarnIf(returnCode == 0, fmt::format("ldms_xprt_connect_by_name failed with code {} \n", returnCode));
+ const auto returnCode =
+ ldms_xprt_connect_by_name(ldms_, hostname, port, NULL, NULL);
+ vtWarnIf(
+ returnCode == 0,
+ fmt::format("ldms_xprt_connect_by_name failed with code {} \n", returnCode)
+ );
}
#endif
@@ -340,9 +343,8 @@ void NodeLBData::outputLBDataForPhase(PhaseType phase) {
void NodeLBData::writeJSONToLDMS([[maybe_unused]] const nlohmann::json& j) {
if (ldms_prev_submission_ == 0) {
ldms_prev_submission_ = MPI_Wtime();
- } else if (
- (MPI_Wtime() - ldms_prev_submission_) * 1000.0 < (double)ldms_milli_freq_
- ) {
+ } else if ((MPI_Wtime() - ldms_prev_submission_) * 1000.0 <
+ (double)ldms_milli_freq_) {
return;
} else {
ldms_prev_submission_ = MPI_Wtime();
@@ -352,7 +354,10 @@ void NodeLBData::writeJSONToLDMS([[maybe_unused]] const nlohmann::json& j) {
const auto returnVal = ldmsd_stream_publish(
ldms_, "vtLBStats", LDMSD_STREAM_JSON, jsonStr.c_str(), jsonStr.length() + 1
);
- vtWarnIf(returnVal == 0, fmt::format("ldmsd_stream_publish returned {}!\n", returnVal));
+ vtWarnIf(
+ returnVal == 0,
+ fmt::format("ldmsd_stream_publish returned {}!\n", returnVal)
+ );
}
#endif
|
Logan590
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
0100832 to
de1449a
Compare
lifflander
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
c49ac24 to
80ed20e
Compare
Contributor
|
@Logan590 Please fix the commit message for ca0cdd3 (just start it with https://github.com/DARMA-tasking/vt/wiki/Pull-Requests#commit-formatting |
cz4rs
reviewed
Jul 21, 2025
cz4rs
reviewed
Jul 21, 2025
cz4rs
reviewed
Jul 21, 2025
Logan590
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
ca0cdd3 to
9637749
Compare
Contributor
|
@lifflander, there are many versions of LDMS. I'm currently using the version 4.3.5 to work on this issue. Does it seem good to you ? |
cz4rs
reviewed
Jul 25, 2025
cz4rs
reviewed
Jul 25, 2025
cz4rs
reviewed
Jul 25, 2025
cz4rs
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
2 times, most recently
from
3105c59 to
a0c37c5
Compare
cz4rs
reviewed
Aug 11, 2025
cz4rs
reviewed
Aug 11, 2025
Comment on lines
+162
to
+171
| if (auto ldms_freq = getenv("VT_LDMS_MILLI_FREQ")) { | ||
| ldms_milli_freq_ = atoi(ldms_freq); | ||
| } | ||
| const auto xPrt = getenv("VT_LDMS_XPRT"); | ||
| const auto auth = getenv("VT_LDMS_AUTH"); | ||
| ldms_ = ldms_xprt_new_with_auth(xPrt, NULL, auth, NULL); | ||
| vtWarnIf(ldms_, "ldms_xprt_new_with_auth failed!"); | ||
|
|
||
| const auto hostname = getenv("VT_LDMS_HOSTNAME"); | ||
| const auto port = getenv("VT_LDMS_PORT"); |
Contributor
There was a problem hiding this comment.
Those variables should be documented along with the LDMS test procedure described earlier. Maybe a wiki page?
cz4rs
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
a0c37c5 to
f1c010d
Compare
cz4rs
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
6968966 to
8146684
Compare
Contributor
|
@JacobDomagala Can you please have a look? |
cz4rs
force-pushed
the
2183-create-ldma-stream-publish-for-phase-data
branch
from
5004ca8 to
ce87f55
Compare
cz4rs
approved these changes
Aug 12, 2025
Contributor
cz4rs
left a comment
cz4rs
left a comment
There was a problem hiding this comment.
Looks good to me in the parts I haven't touched.
lifflander
commented
Aug 12, 2025
Collaborator
Author
lifflander
left a comment
lifflander
left a comment
There was a problem hiding this comment.
Looks good to me!
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #2183