NSF Workshop on Cloud Computing for Cyber-Physical Systems
We invite you to join us at an NSF-sponsored workshop on Cloud Computing for Cyber-Physical Systems held at the Virginia Tech Research Center in Arlington, VA, on March 14th and 15th, 2013. This workshop will bring together leading researchers, practitioners, and stakeholders in the cloud computing and cyber-physical systems (CPS) communities, from diverse universities, companies, and government agencies, towards defining a compelling and novel research agenda at the intersection of cloud computing and CPS, and thus enabling a new generation of safe, secure, reliable, resilient, large-scale systems in areas of vital national interest including energy, transportation, communication, defense, and food and water distribution. We plan to have all attendees help project forward from what is understood about cloud computing and cyber-physical systems today to formulate a cohesive research agenda over the next three to five years to propel a productive integration of cloud computing and CPS and in doing so drive advances in both fields.
Cyber-physical systems feature a tight combination of, and coordination between, the system's computational and physical elements. Increasingly, these are also distributed systems, spanning multiple computing, communication, and physical sub-systems that must be managed rigorously despite the complexity inherent in those ever-larger combinations. Moreover, to overcome limitations of current-generation systems, previously unimagined scales of distribution, resource utilization, and interconnection are needed, which often exceed the current state of the art and raise new research challenges for CPS.
Cloud computing abstracts the delivery of computing and communication services, and with an increasing attention to differentiated quality of service offers potential approaches to address those same research challenges. Specifically, combining cloud computing techniques (such as virtualization, elastic re-configuration, and multi-tenancy of resources) with CPS techniques (such as real-time scheduling, adaptive control, and embedded system design) appears promising to advance the state of the art in multiple research disciplines, and allow previously unachievable systems to be built, deployed, and managed effectively.
The workshop will include selected presentations with new ideas towards (1) identifying limitations of the state of the art in CPS, cloud computing, or relevant application domains, which must be addressed; (2) suggesting new research directions that show potential for overcoming those limitations; and/or (3) how approaches in cloud computing can and should be integrated with approaches in CPS, and vice versa, to achieve new capabilities. The workshop also will involve several working sessions to further collect, expand, and refine ideas related to those central themes. Travel support from NSF will be available.
Douglas C. Schmidt, Vanderbilt University
Chris Gill, Washington University
Jules White, Virginia Tech