Toward a common digital continuum platform for big data and extreme-scale computing (BDEC2)

University Of Tennessee Knoxville, Knoxville TN

Investigators

Abstract

By the end of this decade, the world's store of digital data is projected to reach 40 zetabytes (10 to the power 21 bytes), while the number of network-connected devices (sensors, actuators, instruments, computers, and data stores) is expected to reach 20 billion. While these devices vary dramatically in their capabilities and number, taken collectively they represent a vast "digital continuum" of computing power and prolific data generators that scientific, economic, social, governmental and military concerns of all kinds will want and need to utilize. The diverse set of powerful forces propelling the growth of this digital continuum are prompting calls from various quarters for a next generation network computing platform - a digital continuum platform (DCP) - for creating distributed services in a world permeated by devices and saturated by digital data. But experience shows how challenging the creation of such a future-defining platform is likely to be, especially if the goal is to maximize its acceptance and use, and thereby the size of the community of interoperability it supports. Focusing on the strategically important realm of scientific research, broadly conceived, this project is staging six international workshops (two each in the United States, Europe, and Asia over a two-year period) to enable transnational research communities in a wide range of disciplines to converge on a common DCP to meet this challenge. Building on a decade of leadership in cyberinfrastructure planning, the Big Data and Extreme-scale Computing (BDEC2) community is attacking this problem by pursuing three complementary objectives: 1. Draft a design for a "digital continuum platform" (DCP) to serve as shared software infrastructure for the growing continuum of computing devices and data sources on which future science will rely; 2. Organize and plan an international demonstration of the feasibility and potential of the DCP; and 3. Develop a corresponding "shaping strategy" that addresses all relevant stakeholders and moves the community toward convergence on a standard DCP specification. Thus, the project serves the national interest, as stated by NSF's mission: to promote the progress of science and to secure the national defense. Creating a common digital continuum platform represents a grand challenge problem for the global cyberinfrastructure community. To address this monumental challenge and achieve its objectives, the BDEC2 community is organizing around four distinct but complementary activities: 1. Surveying and analyzing the spectrum of application/workflow needs across diverse research and engineering communities who will use the digital continuum; 2. Developing a reference design for a DCP system architecture that is able to manage the trade offs involved in using a widely shared infrastructure to satisfy diverse application community requirements; 3. Strengthening cooperative and crosscutting efforts among stakeholders (e.g., research communities, commercial vendors, software developers, resource providers) in the "cyber ecosystem" of science and engineering; and 4. Formulating a strategy for building community consensus on a common digital continuum within this same collection of stakeholders. To help researchers converge on critical problems for important user communities, foster and focus collaboration to solve those problems, and better coordinate software research and data sharing, the BDEC2 community process engages both international software research and data science communities. The process also includes inter-meeting working groups (e.g., for application/workflow analysis and DCP architecture). Combined with the work of the international meetings themselves, these working groups are intended to accelerate community-wide discussion and collaborative activities needed to address the multi-dimensional challenges of the emerging digital continuum. By achieving its goals, this project intends to supply the different stakeholder communities with the kind of well-defined vision and consensus building strategy necessary to realize a common, open, and interoperable DCP for digital continuum era. The project actively promotes participation by talented young scientists (up to 15 across the series) drawn from the academic community and with special attention to women and minorities. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.