Neural network using back propagation machine learning model for disk failure prediction Junjie Qian, [email protected]

Input set: 1. read attributes from "configuration" file, format defined; 2. data set format expect as "value1,value2,value3,...,valuen n m value1,...,valuem"(n the input attributes number, m the output attributes number)

Files: "include", all head files; "src" all source files; "obj" all object files; "sample" sample configure file and train/test datasets; "testing" testing files

This work is part of our paper published on IEEE NAS 2015 and later reported by TechRepublic (http://goo.gl/eKYKxk).

  @inproceedings{qian2015p3,
    title={P3: Priority based proactive prediction for soon-to-fail disks},
    author={Qian, Junjie and Skelton, Stan and Moore, Joseph and Jiang, Hong},
    booktitle={Networking, Architecture and Storage (NAS), 2015 IEEE International Conference on},
    pages={81--86},
    year={2015},
    organization={IEEE}
  }

Qian, Junjie, Stan Skelton, Joseph Moore, and Hong Jiang. "P3: Priority based proactive prediction for soon-to-fail disks." In Networking, Architecture and Storage (NAS), 2015 IEEE International Conference on, pp. 81-86. IEEE, 2015.