基于数据挖掘的存储数据纠删码容错方法仿真

在存储数据的纠删码容错中,针对传统存储数据纠删码容错方法容错速度较低的问题,提出一种基于数据挖掘的存储数据纠删码容错方法。采用数据挖掘方法对存储数据纠删码进行重构,计算存储数据纠删码的丢失片段;利用CHR算法对存储数据纠删码进行异构修复,通过建立存储数据纠删码容错模型实现存储数据的纠删码容错。为了验证存储数据纠删码容错方法的有效性,将存储数据纠删码容错方法与传统存储数据纠删码容错方法进行对比,实验结果得出:上述方法与基于流水线的存储数据纠删码容错方法、基于RapidRaid码的存储数据纠删码容错方法、基于非规则LDPC码的存储数据纠删码容错方法的容错速度分别为:286Mbps/s、262Mbps/s、243Mbps/s、232Mbps/s,比较可知,所提方法的容错速度最快,证明了上述方法的优越性。

Fault-Tolerant Simulation of Erasure Codes for Stored Data Based on Data Mining

In this article, a fault-tolerant method for erasure code of storage data based on data mining was put forward. The data mining method was used to reconstruct the storage data erasure code, and the missing segments of storage data erasure code were calculated. CHR algorithm was used to perform the heterogeneous restoration on the storage data erasure code. The fault tolerant for erasure code of storage data was achieved by establishing the fault-tolerant model of storage data erasure code. In order to verify the proposed method, the fault-tolerant method for erasure code of storage data was compared with the traditional method. Simulation results show that the fault-tolerant speed of proposed method is 286 Mbps/s. The fault-tolerant speed of fault-tolerant method for erasure code of storage data based on assembly line is 262 Mbps/s. The fault-tolerant speed of fault-tolerant method for erasure code of storage data based on RapidRaid code is 243 Mbps/s. The fault-tolerant speed of fault-tolerant speed of fault-tolerant method for erasure code of storage data based on the irregular LDPC code is 232 Mbps/s. The comparison shows that the fault-tolerant speed of proposed method is the fastest, which proves the superiority of the proposed method.