A Partition-based Mechanism for Reducing Energy in Phase Change Memory

Phase Change Memory (PCM) is a promising candidate of universal memory due to its non-volatility, high density and bit-alterability. However, high write energy consumption and limited endurance are two major challenges for its adoption in memory hierarchy. To address these challenges, one of the effective methods is to reduce write activities on PCM. This paper targets at PCMbased secondary storage and presents a partition-based mechanism, named Bipartite Matching Write (BMW), to reduce unnecessary write activities on PCM. Based on this mechanism, we propose two kinds of write algorithms called BMW-Greedy and BMW-KM, respectively. Based on greedy algorithm, BMW-Greedy obtains an approximating optimum result by allocating each input sub-block to an optimal available target sub-block. Based on kuhn-munkras algorithm, BMW-KM obtains an optimum result by finding a perfect matching between input sub-blocks and target subblocks. Experimental results show that, the proposed BMWGreedy algorithm achieves 11.29% average energy saving (up to 18.63%) compared to the DCW algorithm and 13.30% average energy saving (up to 20.71%) compared to the Flip-N-Write algorithm; the proposed BMW-KM algorithm achieves 12.80% average energy saving (up to 20.86%) compared to the DCW algorithm and 14.81% average energy saving (up to 22.93%) compared to the Flip-N-Write algorithm.