2022 10 10 Open NEST Developer Video Conference

previous | list | next

Agenda

1. Welcome
2. Review of NEST User Mailing List
3. Project team round
4. In-depth discussion
   • Live Hackathon
   • NEST-GPU pynest interface
   • Placement of extension modules
   • Design Documents
   • e-prop implementation in NEST (postponed to next meeting)

Mailing list

• https://nest-simulator.org/mailinglist

Project team round

Here we discuss topics that need broader attention, for example questions that came up but are outside a single project's scope, larger planned changes/PRs that affect all teams or pending work that is blocked by external factors.

1. Models / NESTML
   • worked on glif implementation with same time-constants (can cause singularity)
   • working on SpiNNaker models
2. NEST GPU
   • work as usual
3. PyNEST
   • nothing to report
4. NEST Desktop
   • work as usual
   • next release planned in October
5. Kernel
   • working on performance bug with partial progress
     • looking into issue with compressed spikes
   • started clang-tidy cleaning, refactoring many points for more modern code
     • needs to be transferred to NESTML, too !!
6. Installation & Infrastructure
   • working on NEST release 3.4
     • run benchmarks on release candidate
7. Documentation
   • nothing to report
8. EBRAINS
   • next EBRAINS relase: feature freeze Oct 15

In-depth discussion

• e-prop implementation in NEST (postponed)
• Placement of extension modules (postponed)
• Design Documents (postponed)

Live Hackathon (Mon 28 – Dec 2, Jülich)

• check mail with details
• fight open PRs
• Dec 1: 10th NEST Initiative Anniversary Talk & Dinner

NEST-GPU pynest interface

• NEST and NEST-GPU are different simulators, design of NEST-GPU motivated by NEST

• both have Python interface, some commands in NEST or NEST-GPU but not the other

• need common set of commands to define models to test on both simulators

• NEST3 pynest interface is well structured, now question

  • how to connect to lower level NEST-GPU functionality
  • how to approach the effort

• NEST-GPU uses Python ctypes library

• NEST side has worked on removing the SLI dependency, i.e. no more ll_api soon

=> exchange of prototype branches and dedicated meeting will be scheduled

Discussed options:

• same hl_api with different ll_api libraries loaded (CPU, GPU, …)
• pybind could make API binding automatic and easier to handle
  • need to look at data types (Vectors, np.arrays, …)
• cython allows to write custom code in the glue