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

{toolchain} intel/2019a (+ HPL 2.3 as test case) #7372

Merged
merged 3 commits into from
Jan 24, 2019

Conversation

boegel
Copy link
Member

@boegel boegel commented Jan 7, 2019

(created using eb --new-pr)

@boegel boegel added the update label Jan 7, 2019
@boegel
Copy link
Member Author

boegel commented Jan 7, 2019

Test report by @boegel
SUCCESS
Build succeeded for 4 out of 4 (4 easyconfigs in this PR)
node3102.skitty.os - Linux centos linux 7.6.1810, Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz, Python 2.7.5
See https://gist.github.com/6cd93ecb41117c6e4552978239d7968b for a full test report.

@hajgato
Copy link
Collaborator

hajgato commented Jan 7, 2019

@boegel change foss to intel in the title?

@boegel
Copy link
Member Author

boegel commented Jan 7, 2019

Test report by @boegel
SUCCESS
Build succeeded for 4 out of 4 (4 easyconfigs in this PR)
node2452.golett.os - Linux centos linux 7.5.1804, Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz, Python 2.7.5
See https://gist.github.com/df60313ef7d4a09e28456a22ee2f7642 for a full test report.

@boegel
Copy link
Member Author

boegel commented Jan 7, 2019

Test report by @boegel
SUCCESS
Build succeeded for 4 out of 4 (4 easyconfigs in this PR)
node2094.delcatty.os - Linux centos linux 7.5.1804, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.7.5
See https://gist.github.com/8da6a5e40a67fdc97475bd45e0cf63d0 for a full test report.

@boegel boegel changed the title [PROPOSAL] {toolchain} foss/2019a (+ HPL 2.3 as test case) [PROPOSAL] {toolchain} intel/2019a (+ HPL 2.3 as test case) Jan 7, 2019
@boegel boegel force-pushed the 20190107114721_new_pr_HPL23 branch from a44af34 to edcaf07 Compare January 7, 2019 15:14
@boegel
Copy link
Member Author

boegel commented Jan 7, 2019

@hajgato Woops, fixed (incl. commit message with git push --force)

@damianam
Copy link
Member

damianam commented Jan 9, 2019

Intel MPI 2019.[0,1] crashes when using most collectives, for most algorithms. That's why I think we should use Intel MPI 2018.4 instead, unless someone finds the underlying problem with the new versions (potentially related to libfabric?)

@akesandgren
Copy link
Contributor

akesandgren commented Jan 10, 2019

IMPI 2019.1 fails on IMB 4.1 Reduce with 1024 tasks (4 bytes) for me, so definitely some problem with it.
At the very least it hangs on that test.

@akesandgren
Copy link
Contributor

And IMPI 2018.3.222 (intel/2018b) works as it should.

So clearly we need to stay away from IMPI 2019.[01]

@boegel boegel added this to the 3.8.1 milestone Jan 12, 2019
…lems Intel MPI 2019 update 1 with large-ish jobs
@boegel boegel force-pushed the 20190107114721_new_pr_HPL23 branch from 84c77e8 to 89b71aa Compare January 12, 2019 18:31
@boegel
Copy link
Member Author

boegel commented Jan 12, 2019

Test report by @boegel
SUCCESS
Build succeeded for 5 out of 5 (5 easyconfigs in this PR)
node3111.skitty.os - Linux centos linux 7.6.1810, Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz, Python 2.7.5
See https://gist.github.com/f0e5d0de69f19c722eb54f39cc4ed72e for a full test report.

@boegel
Copy link
Member Author

boegel commented Jan 12, 2019

Test report by @boegel
SUCCESS
Build succeeded for 5 out of 5 (5 easyconfigs in this PR)
node2617.swalot.os - Linux centos linux 7.6.1810, Intel(R) Xeon(R) CPU E5-2660 v3 @ 2.60GHz, Python 2.7.5
See https://gist.github.com/97454a9f8d0c93719e42a4bd93adb6a3 for a full test report.

@boegel
Copy link
Member Author

boegel commented Jan 12, 2019

Test report by @boegel
SUCCESS
Build succeeded for 5 out of 5 (5 easyconfigs in this PR)
node2001.delcatty.os - Linux centos linux 7.6.1810, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.7.5
See https://gist.github.com/0c9cfedb5c539fda8c653d35c72edcac for a full test report.

@migueldiascosta
Copy link
Member

Test report by @migueldiascosta
SUCCESS
Build succeeded for 5 out of 5 (5 easyconfigs in this PR)
grc-cluster1 - Linux centos 6.10, Intel(R) Xeon(R) CPU E5-2640 0 @ 2.50GHz, Python 2.7.14
See https://gist.github.com/afda356cf5d5dfc0363eb251bc39d2ce for a full test report.

@akesandgren
Copy link
Contributor

Test report by @akesandgren
SUCCESS
Build succeeded for 5 out of 5 (5 easyconfigs in this PR)
b-cn0315.hpc2n.umu.se - Linux ubuntu 16.04, Intel(R) Xeon(R) CPU E5-2690 v4 @ 2.60GHz, Python 2.7.12
See https://gist.github.com/9e604fa098778d39011878742f2b9243 for a full test report.

@boegel boegel changed the title [PROPOSAL] {toolchain} intel/2019a (+ HPL 2.3 as test case) {toolchain} intel/2019a (+ HPL 2.3 as test case) Jan 23, 2019
@boegel
Copy link
Member Author

boegel commented Jan 23, 2019

Detailed test report on re-building all easyconfigs that use intel/2018b as toolchain with the proposed intel/2019a instead is available at https://gist.github.com/boegel/68b542fa564fd370e8b0e7710a320f6f .

Compilation failed for a handful of easyconfigs with cryptic errors (to be figured out), but I don't think any of these should block us from going forward with the current intel/2019a definition (and this was not contested during the last EasyBuild conf call, see https://github.com/easybuilders/easybuild/wiki/Conference-call-notes-20190123#test-report-for-proposed-2019a-common-toolchains).

Summary: 110 successfully installed easyconfigs using intel/2019a (+ 102 more with GCCcore/8.2.0 subtoolchain) vs 3+2 failed installations.

@boegel boegel force-pushed the 20190107114721_new_pr_HPL23 branch from d657b68 to cc7f673 Compare January 24, 2019 09:17
@ocaisa ocaisa merged commit 5e4f36e into easybuilders:develop Jan 24, 2019
@boegel boegel deleted the 20190107114721_new_pr_HPL23 branch January 24, 2019 12:10
@jordiblasco
Copy link
Contributor

jordiblasco commented Feb 27, 2019

Have you seen that there is a new release of Intel MPI (2019U2)?
https://software.intel.com/en-us/articles/intel-mpi-library-release-notes-linux#inpage-nav-3-1
I haven't run any MPI sanity checks yet.

@boegel
Copy link
Member Author

boegel commented Feb 27, 2019

@jordiblasco see #7598

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants