Status of HPC sub-project JIRA Tickets |
Key |
Summary |
Assignee |
Priority |
Status |
Status |
CAS-7300 |
CAS-7276
Test lazy filler option in the MMS context |
Justo
Gonzalez Villalba |
Major |
Open |
4.4,
4.5 |
CAS-7299 |
CAS-7276
create quick-guide for developers to test on MMS |
Sandra
Castro |
Major |
Scheduled |
4.4 |
CAS-7283 |
CAS-7276
test_flagdata seg faults on MMS due to some changes in the VI/VB2
setCacheSize() |
James
Jacobs |
Major |
Open |
blocker |
CAS-7280 |
CAS-7276
Create a new partition schema to distribute the content of each spw and scan
along all subMSs |
Justo
Gonzalez Villalba |
Major |
Scheduled |
4.4 |
CAS-7279 |
CAS-7120
Task setjy does not work anymore on Multi-MS |
Tak
Tsutsumi |
Critical |
Open |
blocker |
CAS-7278 |
CAS-7276
test_setjy needs to be modified to run on MMSs |
Tak
Tsutsumi |
Major |
Open |
critical |
CAS-7277 |
CAS-7276
Improve the make_mmsdata script and convertToMMS class to create MMS data |
Sandra
Castro |
Major |
Scheduled |
DONE |
CAS-7201 |
CAS-7120
Application of online flags in parallel is quite slow for MMS with a large
number of sub-MSs |
Julian
Taylor |
Major |
Scheduled |
4.4 |
CAS-7200 |
CAS-7120
wvrgcal seg.faults when working with MMS |
Dirk
Petry |
Major |
Open |
the
work-around is to avoid WVR spw alone on a subMS. Not a blocker but needs to
be fixed. |
CAS-7199 |
CAS-7276
Need to have functional tests for applycal in order to test it with Multi-MSs |
George
Moellenbrock |
Major |
Open |
Without
sequential tests for applycal, we cannot test the MMS cases. The painful
work-around is to run regressions with MMS, but that takes much longer. It's
critical for 4.4 |
CAS-7189 |
CAS-7276
Tutorials to test the parallelization framework |
Andy
Hale |
Major |
Scheduled |
4.4,4.5 |
CAS-7168 |
CAS-7120
Running flagdata on an sub-MS that has only the WVR spw fails |
Sandra
Castro |
Major |
Resolved |
DONE |
CAS-7167 |
CAS-7120
Merging FEED sub-table should avoid the WVR spw sub-MS |
Sandra
Castro |
Major |
Resolved |
DONE |
CAS-7125 |
parallelization
heirustics for pipeline environment |
Sandra
Castro |
Major |
Scheduled |
Pending
on all fixes |
CAS-7122 |
CAS-7120
NULL MS selection exceptions should be hidden from the client when run in
parallel |
Justo
Gonzalez Villalba |
Major |
Scheduled |
4.4,
4.5 |
CAS-7116 |
Lazy
Filler Completion |
Michel
Caillat |
Major |
Scheduled |
? |
CAS-7110 |
Virtual
image concat along spectral axis (similar to mms) |
Kumar
Golap |
Major |
Scheduled |
? |
CAS-7019 |
Integration
of OpenMPI framework into regular package production |
Julian
Taylor |
Critical |
Scheduled |
4.4 |
CAS-7018 |
Update
openmpi / mpi4casa implementation to respect passed environmental variables |
Justo
Gonzalez Villalba |
Major |
Scheduled |
4.4 |
CAS-7016 |
Option
to open main table in write mode and sub-tables in read mode so they can be
shared |
James
Jacobs |
Critical |
Under
Test |
blocker |
CAS-6775 |
Integration
of mip4casa in development and testing environment |
Julian
Taylor |
Critical |
Scheduled |
4.4 |
CAS-6705 |
CAS-5795
Introduce MPI to communicate and synchronize logging directly at C++ level |
Sandra
Castro |
Major |
Scheduled |
4.4 |
CAS-6680 |
CAS-5795
Parallelization of imaging using MPI framework |
Sanjay
Bhatnagar |
Major |
Scheduled |
4.4,
4.5 |
CAS-6543 |
CAS-6541
mstransform uses as input file the entire MMS instead of the sub-Ms when
running in parallel |
Sandra
Castro |
Critical |
Resolved |
DONE |
CAS-5801 |
CAS-5795
Analysis of MPI implementations (system wise, apart from the python bindings) |
Justo
Gonzalez Villalba |
Major |
Closed |
DONE |
CAS-5800 |
CAS-5795
Analysis of schemes to support MPI in the CASA cluster environment |
Justo
Gonzalez Villalba |
Major |
Resolved |
DONE |
CAS-5795 |
MPI-based
CASA cluster infrastructure |
Sandra
Castro |
Major |
Scheduled |
|
CAS-4860 |
CAS-6541
Multi MS opens all the MS when accessed |
James
Jacobs |
Critical |
Scheduled |
blocker |