Release branch for Horace-4 for IDAAAS - installed on IDAAAS as default - branch Horace-4 prerelease
also separate ticket tracking the release branch
Frequency of standups and sprint reviews - standup mondays only, slots reserved always, new slack channel for updates, sprint review one hour only but reserve 2
Input from Franz (who should be setting tasks for PACE, cf. Hannah Griffin)
PACE-integration CI not working - JW to talk to Adam, also Duc re Spinw disconnect - fill in ticket
Things we said we'd do:
docs
deprecate Horace 3 (after closeout) and items leading to this (=Ross flipchart)
check reading H4 from H3
API (=Radu); document changes where interface can't be provided - another meeting/s with Radu
move ross email to Horace-3 deprecation doc (Pace newsletter)
cut not benchmarked, symmetrise does not use parallel, *fit working, gen working, unary/binary not working, virtual clusters gone: continue delaying some benchmarking, prioritise making parallelised subject to being overall feature complete
filebacked - ok except for tobyfit (special case, not before closeout, multifit ok) - shift pixels awaits assessment of energy focus cut
generic projections - except for instrument projection in tobyfit, but that is closeout (relates to event mode)
python interface - done - mo using to learn
generic minimiser - have interface for other minimisers, could do other grad project moving on from Joshua - TGP, JW to explore
new SQW (includes instruments, detectors to be done, see below) -ok
modelling codes (Euphonic, SpinW, Brille) - Christian says SpinW working, David (was) working with Euphonic FRANZ!
also release versions of Euphonic, SpinW - JW to discuss with Duc - report at next sprint
all tested/testing? - more code coverage needed
Things needing reprioritisation:
Profiling - speed was not identified as a problem in Ross identification of issues. Should we keep this deprioritised? #1490, #1488 data from Toby
SLURM cluster availability after move away from SL7.
New issues:
[ ] #1646 (Alex)
[ ] #1643 (Jacob)
[ ] #1639 (Alex)
[ ] #1637 (Jacob)
[ ] #1630 (Alex)
[ ] #1629 (Alex)
[ ] #1616 (Alex)
[ ] #1613 (unassigned, Alex says may be ill-posed)
Issues in progress:
[ ] #1591 legacy alignment (Alex)
[ ] #1553 (Jacob)
[ ] #1544 hashing separated into excess number of ops (done), hashing algorithm (in progress), save/load hashes (to do) #1636 (Chris)
User bugs:
[ ] #1649 (Alex, Jacob)
Issues postponed till after closeout:
parallel fitting not compatible with SpinW (Duc, ...) - ticket?
detector data import (Duc, Toby-->Chris) - ticket? - related detpar tickets are #1432 #1350 #1298 #1338 #1398
normalised polygons (jacob) - ticket?
help for power users e.g. Radu
[ ] #1596
Issues to be completed before closeout:
NB closeout timetabled for 31 August
backward compatibility (3-->4) and conversion between them #1632 (Jacob)
reduction in duplicate messages from parallel workers
1612 noting that Jacob says cut-sym will work for this (Jacob)
parallel HPC settings, random initial configuration (Jacob) - may be done - #1608 #1607 #1606
[ ] #1611 length of time to take cuts (read/write speed on IDAAAS) - addressible in the long term, 4 solutions (not assigned)
[ ] #1572 (Alex)
[ ] #1561 (Alex)
[ ] #1560 (Alex)
[ ] #1585 (Jacob)
[ ] #1538 (Alex)
[ ] #1536 (Toby)
[ ] #1511 (Alex)
[ ] #1481 (Alex, performance issue)
[ ] #1465 (Alex)
[ ] #1436 (Alex)
[ ] #1519 (Jacob - looks as if this should be closed as workaround has been provided?)
Things to sort out:
Things we said we'd do:
Things needing reprioritisation:
New issues:
Issues in progress:
User bugs:
Issues postponed till after closeout:
Issues to be completed before closeout:
1612 noting that Jacob says cut-sym will work for this (Jacob)