A novel memory compression technique for embedded system

Memory is scarce resource, especially for memory constrained embedded system. More physical memory means more package cost, more power consumption, bigger size, etc. Besides, this issue becomes more serious under the memory requirement trend that the average amount of memory required by applications has been grown 50%~100% every year. When new application's working data sets exceed the physical memory of the embedded devices, it is often impossible for the users to increase the memory as they do to commodity PC. One solution to this problem is memory compression which could increase the available memory of the system. However, frequently performing compression and decompression could degrade the performance of the applications since the decompression occur in the critical path of execution. Based on those observation, this paper presents CMPD(compressed memory with pre-decompression), a novel and efficient software-based memory compression technique. CMPD uses pre-decompression algorithm to hide the decompression and mapping latency. In this paper, CMPD is implemented as a loadable module against Linux 2.6.38 kernel and the evaluation results show that this memory compression scheme could effectively increase the available memory of the system with negligible penalty.