Within the next few years, your smartphone will be able to catalog your belongings, your physical activities, your favorite places and favorite things at these places – even how to get those things. It could even teach you how to fix your car or how to clean your espresso machine. The perception algorithms that make this future possible already exist as prototypes in research labs, but they aren’t accurate enough and are so power-hungry they will rapidly drain your battery. At their annual open house today, Intel Labs Berkeley researchers demonstrated a new project that explores the design of extreme power management technology to enable mobile devices to support always-on mobile perception applications., i.e., image, facial, object, and gesture recognition. The research scope includes techniques and tools for power-usage analysis, automated benchmarking, application tuning, and an application developer toolkit. See more on this and other Intel Labs Berkeley Research projects.
Posts Tagged ‘power management technology’
Innovation@Intel: Power-Aware Perception
How to Achieve 40 Percent Energy Savings in Your Data Center
Today’s data center managers have many techniques to reduce their data center’s energy consumption, but a mixture of techniques is often required to achieve energy saving targets. The key to judging success of any energy reduction strategy is the ability to accurately measure results along each step of the way. Here, Knowledge Center contributor Joe Polastre illustrates how power management technology can be used to achieve up to 40 percent energy savings in your data center.
– Everyone’s
talking about how the 35 to 40 percent energy savings promised by
so-called green data center initiatives can help MIS operations
dramatically reduce both their operating expenses and their
environmental impact. But despite the buzz, there is not much being
said about the most effici…



