Present: Yosef Z, Peter D, Roland H, Bill G, Indrani Banerjee, Gabriele B, Leo W, Ratan Deshpande, Maria BH, Zach E, Atul K
Chair: Roland, Minutes: Atul
Roland: Anything pre-agenda.. None
* presentation topics: Einstein Toolkit Seminar Proposals [https://docs.einsteintoolkit.org/et-docs/Einstein_Toolkit_Seminar_Proposals ] [RH] Additionally Fuka ID by Samuel Tootle.
* midwest relativity meeting 2021 recap [RH, ZE] Leo.. Simpy code. Similar to numpy, Cpy package. Roland: Codes up Einstein notations. Leo: Uses Einstein notations, but uses simpy’s tensors, which is known to be buggy. It is a C code so isn’t going to be as fast.
Gabriele: Meeting worked better for in-person as most attendees were in person. Benefits of online was to attend to only ones interested.
* ET website accessibility automated report https://fae.disability.illinois.edu/summary/17d2ffcd5b2a064e/rc/ [RH] A tool that tests the accessibility of a website for people with disabilities. Number of violations (10) are not that bad. Understanding the violations may take some more evaluations.
* create "community contributed" section of website or wiki (must be easier to edit than right now) [BG] Adding to wikipage? Is tricky. Some other way to maintain these with less barrier would be better. Wolfgang: Not currently doing for RePriMand Leo: NRPy, the new contributor’s name goes into the NRPy page. Kuibit: Github pull request.. Gabriele reviews it and then approves. Roland: News item, where it people add their new work’s description (text and image, parfile etc. and reviewed) Bill will keep it in his todo list and bring it up again in the future.
* next [https://docs.einsteintoolkit.org/et-docs/Release_Process ET release] [SB, YZ, RH] Name: Johnson Steve attending Supercomputing conference. * delay by 1 week to avoid thanksgiving holiday. Will be released in first week of December.
** status update on include candidates: *** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2551/include-reprimand-... Reprimand] Commits for the final review.. Added to the ticket. Some mostly non-issues, some minor changes.. Everything went into the development repo (not public). G and some other constants left for the user to specify.
Not addressed comments about numbers being wrong without giving the offset value. Risk of breaking other things.
Roland: Finished the review. Code is in Excellent state! RePriMand has been added to the thornlist.
*** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2549/inlcude-flrwsolver... FLRWSolver] Postponed to next release.
*** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2538/inclusion-of-kuibi... Kuibit] Yosef: Well documented.. Maybe adding more examples by different users.
Gabriele: Fixed a minor issue about h5py. GB added examples, about 20, himself. Kuibit scripts could be used for gallery examples, GB could help the testers. Documentation is going to be version wise and ET is going to be linking each version wise. All: Great shape and will be added to ET.
** testsuite status: http://einsteintoolkit.org/testsuite_results/index.php Roland started running them a couple of weeks ago. Systems are being tested by Steve, Roland, Gabriele.. Some systems are not being run asther are being replaced by newer machines.
Wolfgang: Adding a standard unit test to the test-suite? Roland: Test-suite runs a parfile and looks at the output. Wolfgang: Asking due to the dependency on Boost. Roland: Adding a new parfile and then checking the output ascii. Would be doable. Could be added as a summer project.
Gabriele: Always have to switch between openmp, or mpi. Is there an easier way to set that up, and benchmark? Roland: could maybe run BBH in some test suite systems and regression test in the compute time. Wolfgang: Parameter space is probably too big. All: But standardization could be useful. Will revisit this proposal again.
** gallery runs *** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2172/test-binary-black-... BBH] [CHC] Ran it halfway, and ongoing. *** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2176/test-binary-neutro... BNS] [AD] Tested completely.
*** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2175/test-single-stable... TOV] [PD] Close to running it. Trying on a new AMD-Epyc cluster in stockholm. Compilation succeeded today. Could also add it to the test suite by running the tests par. May be able to use kuibit.
*** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2173/test-poisson-equat... Poisson] [MG] Ran the simulation, will update the website and ticket soon.
*** [ https://bitbucket.org/einsteintoolkit/tickets/issues/2174/test-multi-patch-s... scalar wave] [BG] Ran it today, plots looking good. Need to update website/ticket. Maybe use kuibit, if time permits.
[https://www.einsteintoolkit.org/tools/unanswered.php unanswered question on mailing list] Not updated: Hee Il Kim’s question.
[ https://bitbucket.org/einsteintoolkit/tickets/issues?status=open&status=... open tickets sorted by update time] Issue #2497. https://bitbucket.org/einsteintoolkit/tickets/issues/2497/illinoisgrmhd-is-i... IllinoisGRMHD schedules the conservative-to-primitive methods in the group AddToTmunu, which is not added if stress_energy_at_RHS is set to ‘no’. Leo and Zach will discuss. https://bitbucket.org/einsteintoolkit/tickets/issues/2576/after-statements-i...
Will be good to add a read-write statement.
[ https://bitbucket.org/einsteintoolkit/tickets/issues?status=new&status=o... tickets ready for review] Not much new. Reviewers needed.
Future chairs and minute takers: 2 Dec: Chair : Yosef, Minutes: Bill
Hello Wolfgang,
Roland: Finished the review. Code is in Excellent state! RePriMand has been added to the thornlist.
Concerning your question about running arbitrary executables other than the main Cactus code for the test. It occurred to me that we have the capability to postprocess test data:
https://www.einsteintoolkit.org/usersguide/UsersGuide.html#verbatim-92
and just above using the:
POSTPROC tool.exe regex-of-files
notation.
Nothing really stops you from having a "dummy" Cactus test that produces an output file and abusing the postproc option to run the actual test executable (and completely ignoring the output file content).
Horrible, but maybe a first try.
Note that there are issues with this. Namely the postproc utility will run on on whichever node ends up running the testsuite driver Perl script. This could be your laptop, the login node of a cluster or a compute node in a cluster (or a launch node in something like a Cray XE system).
There is no guarantee that a login node will be able to run code compiled for the compute nodes (which your tool would be since Cactus *only* knows how to compile for compute nodes, not for the login nodes). This is an issue when cross-compiling. Note that compiling for KNL compute nodes using say a Broadwell login node (or even SkyLake) is cross compiling since the KNL have instructions that the login node cannot execute (and the Intel compiler inserts startup code to main() that checks for them and aborts if they are not found).
Similarly some clusters may just not let you run any compiled code without the equivalent of mpirun (again Crays are good examples for this). Since the testsuite code does not prefix the tools by mpirun, they would fail. Similarly some clusters do not let you execute mpirun on the login nodes.
There are also (to no ones surprise I a guess) issues with executing a secondary tool from within cactus_sim. Namely many MPI implementations do not let you. See eg:
https://www.open-mpi.org/faq/?category=tuning#fork-warning
and note that cactus_sim is an MPI process (it is started via mpirun, it calls MPI_Init() in its main()). Other MPI implementations have similar restrictions.
Yours, Roland
users@lists.einsteintoolkit.org
-
Atul Kedia -
Roland Haas