(1 + ε)-approximate sparse recovery
Eric Price, David P. Woodruff
FOCS 2011
A significant opportunity exists to apply information technology to better understand and more efficiently manage large complex manmade and natural systems. These systems often involve large-scale environmental, monitoring, and control systems, operating in harsh conditions with unreliable and low-bandwidth connections between a variety of heterogeneous devices. While there are many examples of these systems deployed, increasingly, there is a requirement to tightly integrate and couple these systems with more traditional core business applications and underlying infrastructure. This requirement is motivated by the desire for organizations to respond more quickly to new opportunities or threats, to gain deeper insight into the behavior of these systems, and to leverage existing investments in common tools, software stacks, programming models, and infrastructure. Deployment of such systems requires a pervasive messaging infrastructure to enable intercommunication of applications and devices. The publish/subscribe messaging paradigm has a number of characteristics that make it very attractive in this environment. These include scalability, efficiency, decoupling between publishing and subscribing entities, and the ability to integrate a wide variety of devices and systems. In this paper, we explore the application of the publish/subscribe messaging paradigm and a programming model to environmental, monitoring, and control systems, drawing on experience of projects from a variety of industries. © 2010 IBM.
Eric Price, David P. Woodruff
FOCS 2011
Lixi Zhou, Jiaqing Chen, et al.
VLDB
Inbal Ronen, Elad Shahar, et al.
SIGIR 2009
Michael Ray, Yves C. Martin
Proceedings of SPIE - The International Society for Optical Engineering