Compression for Low Power Consumption in Battery-powered Handsets

Abstract

Only summary form given.Java has been introduced to the mobile/wireless handsets and Java enabled handsets are now prevalent in the market, showing their successful launch. A variety of attractive services have been developed and promise a lot of fun to our daily lives, which are however targeted to the high end products that are very expensive. We focus on economic types with similar performance to the high end products. Memory compression is a key technique to achieve the goal and is integrated into a Java runtime environment. We save memory with runtime compression and reduce power by a memory bank partitioning technique (powering off unused memory banks). In memory compression algorithms and the Java runtime environment are studied. The Wilson and Kaplan (WK) algorithm is discussed as in-memory compression and introduced to the Java heap as an alternative of the garbage collection (gc). Experiments show around 10% of speedup and up to 45% of memory saving in the application core part (20.0 % of memory overhead for a user interface). The Java system using the memory compression outperforms its base configuration and will give further chance to reduce memory demand in combination with the garbage collection designed for the Java compressed heap. Experiments show that the in-heap memory compression technique can save more than 50 % (up to 60 %) of the heap demand and that half of the memory banks for the Java heap may never be turned on. 50 % of the power consumption (memory leakage current) can be saved. The hardware compressor and decompressor are currently ten thousand times faster than the software counterparts. The time overhead could be negligible by introducing the hardware compressor and decompressor. Hardware/software codesign is an additional key to achieve our goal. The contribution of the runtime memory and heap compression is threefold: small memory demand, lower power consumption, and negligible time overhead in battery-powered handsets. Future work includes several issues unsolved in the Java memory compression: studies of integration of CLDC Hot Spot implementation and extension of the Java virtual machine to the traditional virtual machine.