Elsa Gonsiorowski, PhD

improve testing

5

There are a number of task that should be done to improve the testing: - [ ] improve the setup of the Travis tests. This should involve using multiple [build...

statistics API

1

ROSS should define an API for model statistics.

would result in one less mpi_allreduce?? mpi datatype might be needed.

would be nice!

PE periodic function

3

API for a callback for a PE function that gets called every GVT (or so). Initially introduced in e2309b79a6605864099e6acb6e06e87aa27f0515 (Jan 2011) Removed by me in 443bfd2b7a393a40b9bbf04fb2f3257f16150e95 (Feb 2014). Should we...

Model Event-Efficiency Statistic

2

a statistic that calculates, on average, how many events are created per event processed. Can be used during model development to understand if a model has an unstable event population.

pe types

1

I believe this "feature" should be deprecated and removed. PE types are used to do logging and statistics reductions... and only in the propagation, srw, and tlm models.

Currently we use either a hash table or an AVL tree for remote event tracking during optimistic simulation. However, the AVL tree is hacked in underneath the hash table implementation....

We should consider: https://exaworks.org/psi-j-python/ It’s a portable submission interface for job schedulers in python. Maybe this will be helpful in SCR.

tests: compilers

1

Should test all the possible compilers: intel, gcc, clang, pgi, etc.