Page History

Mid-term planning (~6 hours*2)

IM_calc performance improvement: only calculating 090 and 000 when only geom is needed

Done

performance improved by 50% (58s vs 38s) by adding a geom flag to read_wavfeform so that waveform object will be reassembled to have only 2 components. Needs a better way as read_waveform should not be aware of what geom is.

Subduction cybershake (v19p4) from NZFM (just subduction-interface)

• How well does SRF generation (only ~10) work? : Does it work?
• VM generation: Plotting
• Estimation work?
• Present numbers before running

Automated workflow improvements

• Fixes mgmt db locking issue
• Estimation models are only loaded once -> increasing submit throughput. HPC queue can easily be kept at full capacity now.

Subduction srf gen code

Jason

Claudio

James

SRF : Slow, memory intensive.

VM: Fixed, verification in progress.

Estimation: Estimation complete, comparison to be done once simulation completes.

HF finding bugs.

BB/IMCalc yet to be done (status unknown)

Done

Code acquired from Rob and compiled.

Yet to be verified.

1. Disply Physical core hours  instead of visual core hours
2.  Both Maui/Mahuika  user core hour usage old data collection + plot
3. Both Maui/Mahuika  total core hour usage old data collection + plotIM_calc performance improvement: only calculating 090 and 000 when only geom is needed

1-3 Done

http://hypocentre:8050/4. performance improved by 50% (58s vs 38s) by adding a geom flag to read_wavfeform so that waveform object will be reassembled to have only 2 components. Needs a better way as read_waveform should not be aware of what geom is.

Epsilon calculation

Ground motion integration

List of items and show the progress

Login, but no signup

Done (except GM selection)

e2e_test:

1. Run validation through existing e2e tests
2. source generation in validation 
3. ensure that different GMSim versions can be handled

1. In progress, resulted in mgmt db locking issue, which had to be fixed first. Ran 626/267 validation cases, just requires some minor fixes/improvements.

PGA/PGV scaling

Probability calculations for Jessee 17/ Zhu16 general
Probability/susceptibility calculations for more other models

Investigate shp files

Calculation is done

Workflow needs to be verified

(James/Jason)

.shp file for next sprint

1. Version number for GMSim workflow can be any string
2. Neural network needs training
3. Mgmt DB locking issue
4. Merge Ts
5. Mahuika queue check before submission
6. Reduce extents for VMs out of (165,180) longitude range

James

Claudio

1. To do

2-5. Done

6. To be verified