|
Under Limited Power and Memory Resources |
We envision a future in which users will rely on a mobile computing device to access information as well as fulfill their computing needs. While some tasks will be performed locally on the mobile device, others tasks will be performed remotely on a computing infrastructure provided much like any other public utility. While local computing must be carried out in a manner which delivers performance under the constraints of limited memory and power resources, remote computing must be carried out to limit communication while providing security and privacy to the user. After the discussion of the research issues, I will present some of our recent results on optimizing performance as well as memory and power usage on a mobile device. First I will show how compact code can be executed to deliver performance through proper instruction set and microarchitectural support. Second I will show how compacted narrow width data, prevelant in multimedia codes, can be effectively manipulated through new compiler algorithms and instruction set support.