MesoNet: remote/cloud test suite

cloud setup notes need the actual environment + command. tiny local/remote comparison before the long run, and keep compilation time separate

recorded change: Google Cloud GPU instances and an initial testing suite were set up. This entry provides an execution direction, not sufficient detail to recreate the cloud machine or calculate cost.

run manifest to recover: machine/GPU type, image, driver/toolkit, Python environment, simulator backend, dataset version, seeds, command and output path. Keep setup/compile time separate from steady-state run time.

comparison plan: run the same small baseline locally and remotely before scaling the workload. Preserve scientific parameters while changing the execution environment, then compare output statistics as well as wall time.

• review questions: which backend actually used the GPU, and were data transfer, logging and compilation included in runtime? No hardware specification or completed speedup is inferred from “GPU instance.”

• References saved in this journal section

• https://www.run.ai/guides/cloud-deep-learning/google-cloud-gpu
• https://console.cloud.google.com/compute/instancesAdd?invt=AbkPjg&project=nmctests
• https://www.google.com/search?q=nvidia+l4&oq=nvidia+l4&sourceid=chrome&ie=UTF-8
• https://brian2cuda.readthedocs.io/en/latest/introduction/index.html
• https://github.com/brian-team/brian2cuda

Iterative development