Power and Performance Analysis of PDA Architectures

As Personal Digital Assistant (PDA) usage increases, so does the demand for better performance while adhering to strict power consumption limitations. As of the time of this study, the majority of PDA usage involves user-level applications which are not yet as computationally intensive as their desktop PC workstation counterparts. Because of that, it seems inappropriate to apply common benchmarks to these architectures, as it would not yield a particularly relevant analysis of how well the performance of these machines meets with current needs. Accordingly, new benchmarks specifically targetting current usage patterns must be devised. By analyzing the performance patterns and power usage footprint of codes that specifically target common user tasks and hardware subsystems (such as LCD display, audio, etc) one can isolate areas of weakness in present architectures.