Exploring the creation of a network of cloud-based experiments

[sgbaird]

@blaiszik, you had expressed interest in this, and I'm curious to get more of your thoughts. What would you be interested in implementing? To spur ideas, maybe you could also take a look at the notebooks, specifically the MQTT section.

Related:

• What does access restriction look like for a scaled up cloud-accessible lab? #91
• https://github.com/sparks-baird/self-driving-lab-demo/discussions/categories/data-and-access-management

[sgbaird]

See Figure 4 from:

Stach, E.; DeCost, B.; Kusne, A. G.; Hattrick-Simpers, J.; Brown, K. A.; Reyes, K. G.; Schrier, J.; Billinge, S.; Buonassisi, T.; Foster, I.; Gomes, C. P.; Gregoire, J. M.; Mehta, A.; Montoya, J.; Olivetti, E.; Park, C.; Rotenberg, E.; Saikin, S. K.; Smullin, S.; Stanev, V.; Maruyama, B. Autonomous Experimentation Systems for Materials Development: A Community Perspective. Matter 2021, 4 (9), 2702–2726. https://doi.org/10.1016/j.matt.2021.06.036.

Schematic showing the expected exponential increase of the speed of research as AE is further developed

We see a progression from connected AE systems to locally integrated systems, and finally to globally integrated systems. At a critical (or tipping) point, integration will create network effects that multiply the contribution of individual research nodes, greatly increasing research speed. Global integration and reduced cost will exponentially impact the access of researchers to AE systems. By leveraging network effects from data sharing and interpretation, and from the community-driven approach to scientific investigation, we anticipate solutions to currently intractable problems.

An interesting thought here is that for a globally integrated system, it can be evaluated whether the travel time and cost of transferring samples between SDLs outweighs the cost of creating the sample on a similar SDL.

[sgbaird]

Nice video from Strateos three years ago that captures the motivation: https://youtu.be/L1UgdoP2aeg