Skip to content
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.

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

ADTM KNL build on mutrino to use OMP_NUM_THREADS=2 #3278

Merged
merged 1 commit into from
Aug 11, 2018

Conversation

fryeguy52
Copy link
Contributor

@trilinos/framework @bartlettroscoe

Description

Change atdm KNL build on mutrino to use OMP_NUM_THREADS=2 previously it was set to 16. This change in local testing of the panzer test suite caused the tests to run in about 1/2 of the time.

Motivation and Context

The panzer tests were being run in about 1/2 the time in the Empire KNL build on mutrino. This is to get the ATDM build to finish in a comparable time

How Has This Been Tested?

I tested this locally by runnig tests with the two different values for OMP_NUM_THREADS

with OMP_NUM_THREADS=2 Total Test time is 132.80 sec

ctest -j8 -R PanzerAdaptersSTK_CurlLaplacianExample
Test project /lscratch1/jfrye/SRC_AND_BUILD/BUILD
    Start 133: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-4
    Start 132: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-3
1/5 Test #132: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-3 ...   Passed   44.96 sec
    Start 131: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-2
2/5 Test #133: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-4 ...   Passed   77.43 sec
    Start 130: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-1
3/5 Test #131: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-2 ...   Passed   68.40 sec
    Start 129: PanzerAdaptersSTK_CurlLaplacianExample
4/5 Test #130: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-1 ...   Passed   41.77 sec
5/5 Test #129: PanzerAdaptersSTK_CurlLaplacianExample .........................   Passed   16.15 sec

100% tests passed, 0 tests failed out of 5

Subproject Time Summary:
Panzer    = 994.88 sec*proc (5 tests)

Total Test time (real) = 132.80 sec

env | sort|grep -i omp_
OMP_NUM_THREADS=2

with OMP_NUM_THREADS=16 total test time is 407.84 sec

ctest -j8 -R PanzerAdaptersSTK_CurlLaplacianExample
Test project /lscratch1/jfrye/SRC_AND_BUILD/BUILD
    Start 133: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-4
    Start 132: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-3
1/5 Test #133: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-4 ...   Passed  269.91 sec
    Start 131: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-2
2/5 Test #132: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-3 ...   Passed  315.79 sec
    Start 130: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-1
3/5 Test #131: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-2 ...   Passed  104.72 sec
    Start 129: PanzerAdaptersSTK_CurlLaplacianExample
4/5 Test #130: PanzerAdaptersSTK_CurlLaplacianExample-ConvTest-Quad-Order-1 ...   Passed   67.46 sec
5/5 Test #129: PanzerAdaptersSTK_CurlLaplacianExample .........................   Passed   30.81 sec

100% tests passed, 0 tests failed out of 5

Subproject Time Summary:
Panzer    = 3154.75 sec*proc (5 tests)

Total Test time (real) = 407.84 sec

env | sort|grep -i omp_
OMP_NUM_THREADS=16

updade mutrino envronment script to use OMP_NUM_THREADS=2 instead of
OMP_NUM_THREADS=16.  In local testing this makes the test suite for
panzer run about twice as fast
@fryeguy52 fryeguy52 added client: ATDM Any issue primarily impacting the ATDM project ATDM Config Issues that are specific to the ATDM configuration settings labels Aug 10, 2018
@bartlettroscoe bartlettroscoe added the stage: in progress Work on the issue has started label Aug 10, 2018
@trilinos-autotester
Copy link
Contributor

Status Flag 'Pre-Test Inspection' - Auto Inspected - Inspection Is Not Necessary for this Pull Request.

@trilinos-autotester
Copy link
Contributor

Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects:

Pull Request Auto Testing STARTING (click to expand)

Build Information

Test Name: Trilinos_pullrequest_gcc_4.9.3

  • Build Num: 1330
  • Status: STARTED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.9.3
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_gcc_4.8.4

  • Build Num: 1024
  • Status: STARTED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.8.4
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_intel_17.0.1

  • Build Num: 574
  • Status: STARTED

Jenkins Parameters

Parameter Name Value
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Using Repos:

Repo: TRILINOS (fryeguy52/Trilinos)
  • Branch: atdm-KNL-mutrino
  • SHA: 6c02877
  • Mode: TEST_REPO

Pull Request Author: fryeguy52

@trilinos-autotester
Copy link
Contributor

Status Flag 'Pull Request AutoTester' - Jenkins Testing: 1 or more Jobs FAILED

Note: Testing will normally be attempted again in approx. 2 Hrs 30 Mins. If a change to the PR source branch occurs, the testing will be attempted again on next available autotester run.

Pull Request Auto Testing has FAILED (click to expand)

Build Information

Test Name: Trilinos_pullrequest_gcc_4.9.3

  • Build Num: 1330
  • Status: FAILED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.9.3
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_gcc_4.8.4

  • Build Num: 1024
  • Status: PASSED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.8.4
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_intel_17.0.1

  • Build Num: 574
  • Status: PASSED

Jenkins Parameters

Parameter Name Value
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde


CDash Test Results for PR# 3278.

@mhoemmen mhoemmen added the AT: RETEST Causes the PR autotester to run a new round of PR tests on the next iteration label Aug 10, 2018
@bartlettroscoe
Copy link
Member

FYI: For details on the above PR testing failure see #3276 (comment).

Copy link
Member

@bartlettroscoe bartlettroscoe left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good.

@@ -51,7 +51,7 @@ if [ "$ATDM_CONFIG_COMPILER" == "INTEL" ] && [ "$ATDM_CONFIG_KOKKOS_ARCH" == "HS
elif [ "$ATDM_CONFIG_COMPILER" == "INTEL" ] && [ "$ATDM_CONFIG_KOKKOS_ARCH" == "KNL" ]; then
module use /projects/EMPIRE/mutrino/tpls/knl/modulefiles
export SLURM_TASKS_PER_NODE=16
export OMP_NUM_THREADS=16
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wow, I can't believed I messed that when I was looking at this. Going for 16 down to 2 threads per process should make a huge difference.

@trilinos-autotester
Copy link
Contributor

Status Flag 'Pull Request AutoTester' - User Requested Retest - Label AT: RETEST will be reset after testing.

@trilinos-autotester
Copy link
Contributor

Status Flag 'Pull Request AutoTester' - Testing Jenkins Projects:

Pull Request Auto Testing STARTING (click to expand)

Build Information

Test Name: Trilinos_pullrequest_gcc_4.9.3

  • Build Num: 1336
  • Status: STARTED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.9.3
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_gcc_4.8.4

  • Build Num: 1030
  • Status: STARTED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.8.4
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_intel_17.0.1

  • Build Num: 579
  • Status: STARTED

Jenkins Parameters

Parameter Name Value
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Using Repos:

Repo: TRILINOS (fryeguy52/Trilinos)
  • Branch: atdm-KNL-mutrino
  • SHA: 6c02877
  • Mode: TEST_REPO

Pull Request Author: fryeguy52

@trilinos-autotester
Copy link
Contributor

Status Flag 'Pull Request AutoTester' - Jenkins Testing: all Jobs PASSED

Pull Request Auto Testing has PASSED (click to expand)

Build Information

Test Name: Trilinos_pullrequest_gcc_4.9.3

  • Build Num: 1336
  • Status: PASSED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.9.3
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_gcc_4.8.4

  • Build Num: 1030
  • Status: PASSED

Jenkins Parameters

Parameter Name Value
COMPILER_MODULE sems-gcc/4.8.4
JENKINS_BUILD_TYPE Release
JENKINS_COMM_TYPE MPI
JENKINS_DO_COMPLEX OFF
JENKINS_JOB_TYPE Experimental
MPI_MODULE sems-openmpi/1.8.7
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde

Build Information

Test Name: Trilinos_pullrequest_intel_17.0.1

  • Build Num: 579
  • Status: PASSED

Jenkins Parameters

Parameter Name Value
PULLREQUESTNUM 3278
TEST_REPO_ALIAS TRILINOS
TRILINOS_SOURCE_BRANCH atdm-KNL-mutrino
TRILINOS_SOURCE_REPO /~https://github.com/fryeguy52/Trilinos
TRILINOS_SOURCE_SHA 6c02877
TRILINOS_TARGET_BRANCH develop
TRILINOS_TARGET_REPO /~https://github.com/trilinos/Trilinos
TRILINOS_TARGET_SHA e91cdde


CDash Test Results for PR# 3278.

@trilinos-autotester trilinos-autotester removed the AT: RETEST Causes the PR autotester to run a new round of PR tests on the next iteration label Aug 11, 2018
@trilinos-autotester
Copy link
Contributor

Status Flag 'Pre-Merge Inspection' - SUCCESS: The last commit to this Pull Request has been INSPECTED AND APPROVED by [ bartlettroscoe mhoemmen ]!

@trilinos-autotester
Copy link
Contributor

Status Flag 'Pull Request AutoTester' - AutoMerge IS ENABLED, but the Label AT: AUTOMERGE is not set. Either set Label AT: AUTOMERGE or manually merge the PR...

@bartlettroscoe bartlettroscoe merged commit f2b6ef2 into trilinos:develop Aug 11, 2018
@bartlettroscoe bartlettroscoe removed the stage: in progress Work on the issue has started label Aug 11, 2018
tjfulle pushed a commit to tjfulle/Trilinos that referenced this pull request Dec 6, 2018
updade mutrino envronment script to use OMP_NUM_THREADS=2 instead of
OMP_NUM_THREADS=16.  In local testing this makes the test suite for
panzer run about twice as fast
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ATDM Config Issues that are specific to the ATDM configuration settings client: ATDM Any issue primarily impacting the ATDM project
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants