Welcome to PEARC20! PEARC20’s theme is “Catch the Wave.” This year’s theme embodies the spirit of the community’s drive to stay on pace and in front of all the new waves in technology, analytics, and a globally connected and diverse workforce. We look forward to this year’s PEARC20 virtual meeting, where we can share scientific discovery and craft the future infrastructure.
The conference will be held in Pacific Time (PT) and the times listed below are in Pacific Time.
The connection information for all PEARC20 workshops, tutorials, plenaries, track presentations, BOFs, Posters, Visualization Showcase, and other affiliated events, are in the PEARC20 virtual conference platform, Brella. If you have issues joining Brella, please email pearcinfo@googlegroups.com.
We are pleased to welcome a panel of diverse scientists and engineers to present and converse on the challenges facing modern data-driven science workflows. This discussion will prove informative and useful to all aspects of the cyberinfrastructure community: computing, software, networks, funding agencies, researchers, and community organizers.
Organizers: Barr von Oehsen (Rutgers University; Advanced Research Computing) Scotty Strachan (University of Nevada, Reno; Cyberinfrastructure)
Panelists: Jayne Bormann (AlertWildfire camera network) Ivan Seskar (COSMOS wireless testbed) Robert Casey (IRIS seismology data services) David Coyle (USGS Next-gen Water Observing System) Kobus Van der Merwe (POWDER wireless testbed)
Session abstract: Technology evolution in academic science and engineering research has arguably created a new, disruptive workflow that is challenging traditional cyberinfrastructure models designed to solve "elephant flow" and "HPC" problems. These new "Edge-to-Edge" (E2E) workflows typically begin with distributed Internet of Things (IoT) data acquisition, traverse sometimes vast heterogeneous networks, and ultimately result in a spectrum of data products delivered in near-real-time to diverse end stakeholders. E2E is a key workflow type in emerging convergence research, as new observational data with rapid delivery become useful across multiple disciplines, and IoT devices are increasingly affordable for the individual researcher. Dominant domains of E2E workflows include environmental science (public hazards, ecosystem health, and resources management) and smart cities (technology testbeds, social services, and infrastructure planning). A recent "Dear Colleague" RFI letter from NSF (20-015) recognized the need for data-focused, ecosystem-scale CI services and capabilities that transcend disciplines and strive to make data available for seamless interoperability. Traditional academic CI at both the national and institutional scale is highly biased towards single lumps of data and scalable batch computing. E2E workflows represent a "ground zero" for this problem, as IoT data integration and automation is highly complex. The CI community needs to initiate a purposeful pivot in focus, expertise, and practices development to meet these impending needs and the resulting deluge of data flows and varieties. Our panel highlights some mature E2E projects and workflows, and poses challenging questions to both the domain science and CI provider audiences.
