Abstract: Abstract: In the distributed storage system, disk space utilization is generally used to determine the load balance of each storage nodes. When given the equal disk space utilization in each nodes, the balance of storage load is achieved in whole distributed storage system. However, in practical, storage nodes are relatively low in disk IO speed and reliability, which is a bottleneck for the performance of data IO in whole storage system. Therefore in heterogeneous distributed storage system and specially the system which has great differences in disk IO rate and reliability of each storage nodes, the rate of data IO is necessarily limited, when disk space utilization is the only evaluation criteria of storage load balance. A new idea based on read-write efficiency is proposed about measuring the storage load balance in the distributed storage system. According to the definition of storage entropy given by the theory of load balance and entropy, a kind of load balance algorithm based on storage entropy (SE) is proposed. With system load and single node load determination as well as load shifting, this algorithm achieves the quantitative adjustment for storage load of the distributed storage system. Compared and experimented with load balance algorithm based on disk space utilization (DU), it is verified that the proposed algorithm have the advantage of balance for storage load in the distributed storage system. It also effectively restrain the system load unbalance and improve the overall efficiency of reading and writing of the distributed storage system.