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

Merging master back into RC #231

Merged
merged 286 commits into from
Jun 24, 2020
Merged

Merging master back into RC #231

merged 286 commits into from
Jun 24, 2020

Conversation

JulianKunkel
Copy link
Collaborator

No description provided.

JulianKunkel and others added 30 commits October 6, 2018 16:54
Make write verification work even without read test
This patch enables correct compilation on both MacOS and Linux using only
POSIX.1-2008-compliant C with XSI extensions.

Specifically, POSIX.1-2008 is the minimum version because we use strdup(3);
explicit XSI is required to expose putenv from glibc.
mkdir/access syscalls used in PrependDir() should call their equivalent
implemented by the backend.
ime_native_mkdir() and ime_native_rmdir() are available starting
from IME native API version 1.3.
The parser now supports concurrent parsing of all plugin options.
Moved HDF5 collective_md option into the backend as an example.
Example: ./src/ior -a dummy --dummy.delay-xfer=50000
IME: add support of mkdir/rmdir into the IME backend
IOR uses the GetTimeStamp() wrapper to allow gettimeofday() to be used for timings instead of MPI_Wtime() to calculate throughput. This commits adds the same logic to mdtest timings.
mdtest: Use IOR's GetTimeStamp() instead of MPI_Wtime()
Currently, calculation for barriers case looks a bit
suprising, for example, Min stats is from min of
Max value of number of mpi proc, this makes Min
and Max very similar in our testing thus we get
a small Std value too.

I am not a MPI expert, but question is why we
don't use more nature calcuation which calculate
all results from different iterations and MPI proc?

Signed-off-by: Wang Shilong <wshilong@ddn.com>
mdtest: fix calculation for barriers
Moved _XOPEN_SOURCE to config.h; resolves #93
Changed the parser to fix #98.  Does _not_ yet have the HDF5 backend args updated.  They were reverted to allow this to make the 3.2 release and should be re-applied as a separate commit.
Dry-run option to test the configuration / scripts.
Signed-off-by: Mohamad Chaarawi <mohamad.chaarawi@intel.com>
Mark Nelson and others added 29 commits March 10, 2020 10:19
Signed-off-by: Mark Nelson <mnelson@redhat.com>
[RFC] src/aiori-CEPHFS: New libcephfs backend
Display host names when using the -j argument (outlierThreshold) so
that IOR could help to better identify slow nodes:

WARNING: for client02, task 6, write elapsed transfer time is 65.738326
Display outlier host names
Signed-off-by: Mohamad Chaarawi <mohamad.chaarawi@intel.com>
Signed-off-by: Mohamad Chaarawi <mohamad.chaarawi@intel.com>
DFS driver update to share DFS mount on init
… Functionality should presumably be moved into AIORI backends.
Signed-off-by: Mohamad Chaarawi <mohamad.chaarawi@intel.com>
TODO: Check potential issue when using IOR scripts with different backends.
Signed-off-by: Mark Nelson <mnelson@redhat.com>
Signed-off-by: Mark Nelson <mnelson@redhat.com>
Signed-off-by: Mark Nelson <mnelson@redhat.com>
src/aiori-CEPHFS: Option warnings, shutdown checks, and cmount nullification
- init/fini can be made multiple times
- reset tunables on fini so they can change on a re-init

Signed-off-by: Mohamad Chaarawi <mohamad.chaarawi@intel.com>
DFS: allow multiple IO phases when ior_run() is called
mdtest: allocate aligned buffers to support DirectIO
Fixes a typo in initial posix_memalign commit for DirectIO support.
mdtest: use correct number of bytes for posix_memalign of read buffer
@JulianKunkel JulianKunkel merged commit 2a17f3e into rc Jun 24, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.