Towards a Software Development Framework for Interconnected Science Ecosystems

Conference · Sun Jan 01 00:00:00 EST 2023

OSTI ID:2000417

Malviya, Addi Thakur ^[1]; Hitefield, Seth ^[1]; McDonnell, Marshall  ^[1]; Wolf, Matthew  ^[1]; Archibald, Richard  ^[1]; Drane, Lance ^[1]; Roccapriore, Kevin  ^[1]; Ziatdinov, Maxim  ^[1]; McGaha, Jesse ^[1]; Smith, Robert  ^[1]; Hetrick, John ^[1]; Abraham, Mark ^[1]; Yakubov, Sergey ^[1]; Watson, Gregory  ^[1]; Chance, Ben ^[1]; Nguyen, Clara ^[1]; Baker, Matthew ^[2]; Michael, Robert ^[3]; Arenholz, Elke ^[4]; Mintz, Ben ^[1]

---

1. ORNL
2. Voltron Data
3. Roche Sequencing Solutions
4. Lawrence Berkeley National Laboratory (LBNL)

The innovative science of the future must be multi-domain and interconnected to usher in the next generation of “self-driving” laboratories enabling consequential discoveries and transformative inventions. Such a disparate and interconnected ecosystem of scientific instruments will need to evolve using a system-of-systems (SoS) approach. The key to enabling application integration with such an SoS will be the use of Software Development Kits (SDKs). Currently, SDKs facilitate scientific research breakthroughs via algorithmic automation, databases and storage, optimization and structure, pervasive environmental monitoring, among others. However, existing SDKs lack instrument-interoperability and reusability capabilities, do not effectively work in an open federated architectural environment, and are largely isolated within silos of the respective scientific disciplines. Inspired by the scalable SoS framework, this work proposes the development of INTERSECT-SDK to provide a coherent environment for multi-domain scientific applications to benefit from the open federated architecture in an interconnected ecosystem of instruments. This approach will decompose functionality into loosely coupled software services for interoperability among several solutions that do not scale beyond a single domain and/or application. Furthermore, the proposed environment will allow operational and managerial inter-dependence while providing opportunities for the researchers to reuse software components from other domains and build universal solution libraries. We demonstrate this research for microscopy use-case, where we show how INTERSECT-SDK is developing the tools necessary to enable advanced scanning methods and accelerate scientific discovery.

• Research Organization: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC05-00OR22725
• OSTI ID: 2000417
• Resource Relation: Conference: Smoky Mountains Computational Sciences and Engineering Conference (SMC) - Kingsport, Tennessee, United States of America - 8/24/2022 8:00:00 AM-8/25/2022 8:00:00 AM
• Country of Publication: United States
• Language: English