• DREAM
• Program Context Based Prediction
• User Behavior Based Prediction
• Energy-Relibility of Devices
• Sensor-Enabled Devices
• DREAM Team
• Sponsors

User-Behavior-Based Prediction

User interaction provides valuable contextual hints that can be used to determine I/O access patterns in applications. Correlating user interaction, such as mouse clicks, with an application's execution context allows for context-aware power management (CPM), resulting in significant improvements over previous power management schemes. Recent work has shown the potential of mouse clicks for adapting the power mode of the wireless network interface to reflect user interaction, optimizing both the energy and bandwidth usage of the wireless network interface. Continuing work includes the application of machine learning to more complex interaction patterns and the uses of interaction-based power management in the broader context of resource allocation and management.

Papers

• IADVS: On-Demand Performance for Interactive Applications
  Mingsong Bi, Igor Crk, Chris Gniady
  In the 16th IEEE International Symposium on High-Performance Computer Architecture (HPCA-16). 2010
  

• Interaction-aware energy management for wireless network cards.
  Igor Crk, Mingsong Bi, and Chris Gniady.
  In Proceedings of the 2008 ACM SIGMETRICS international conference on Measurement and modeling of computer systems. 2008.

• Context-Aware Mechanisms for Reducing Interactive Delays of Energy Management in Disks.
  Igor Crk and Chris Gniady.
  In Proceedings of the USENIX 2008 Annual Technical Conference. 2008.

• CPM: Context-Aware Power Management in WLANs.
  Fahd Albinali and Chris Gniady.
  In Eighteenth Innovative Applications of Artificial Intelligence Conference (IAAI) July 2006.