Jonathan Karr

Can be done once biosimulators/biosimulators_utils#22 is addressed. Low priority because few simulation tools and few existing SED-ML files take advantage of these features.

Improve handling of large simulation projects

2

Currently, large projects can partly be handled: - The file upload limit for the API and COMBINE API allow COMBINE archives up to 1GB (larger can be uploaded via URLs)...

Add UI to edit simulation experiments (SED-ML documents in COMBINE archives)

8

@jmdetloff here's some starting points ## Goal A UI which enables users to quickly simulate models under alternate conditions (e.g., knockout a gene, change the initiation concentration of a drug,...

This will further reduce the load on the API by shifting it to the dispatch service. Todo - [ ] Move `_getRunSummary` in api to dispatch service - [ ]...

This will further reduce the load on the API by shifting it to the dispatch service. This will also - Ensure that runs can be published quickly once finished (because...

Examples: - GPU - CUDA level - SSE level See also: https://github.com/CBC-UCONN/Slurm_Feature_Constraints

Errors aren't being raised when simulation results fail to upload. This results in corrupt data, where HDF5 datasets are defined the HSDS, but have empty values. This is due to...

Currently, projects are associated with the Auth0 organization(s) that their owner belongs to. Ideally, at the time of publication (a) users would indicate the organization they want to post their...

Improve token proxying by dispatch service

1

Currently, the dispatch service captures the intended owner of a project for the submitted simulation run. Instead, the dispatch service could capture the user's token and pass this along to...