Research On Data Reliability And System Energy Saving In RAID

With the rapid development of communication technology and network technology,digital information is exponentially explosive growth and the data storage technology is facing tremendous challenges.Therefore,people pay more and more attention to the reliability of data in storage system and the energy consumption of storage system.Although the traditional RAID(Redundant Arrays of Independent Drives,RAID)storage technology can effectively improve the reliability of stored data to a certain extent,with the advent of the information age,the amount of data needed to be stored is increasing,the number of disks needed in RAID storage system is also increasing,and the probability of disk damage is also increasing.Therefore,the traditional RAID technology has become increasingly difficult to ensure the reliability of data on disk,and the energy consumption caused by the large number of disks running in RAID storage system can't be ignored.In this paper,erasure coding technology is used in RAID storage system to improve the reliability of data.At the same time,energy consumption in RAID storage system is considered.Energy consumption of storage system is reduced by optimizing data layout.The main contents of this paper are as follows:1.The development history and principle of RAID storage technology are summarized.The basic principle and classification of erasure code technology are introduced,and the related energy-saving technologies in disk array are also introduced.The research idea of combining error-tolerant technology with energy-saving technology is put forward,which can not only ensure the reliability of data in storage system,but also reduce the energy consumption of the system.2.Based on the existing STAR erasure codes,this paper designs an array erasure codes called CPC(Cross Parity Check,CPC)codes,which can tolerate the simultaneous failure of three disks in RAID storage system.In this paper,the encoding method of CPC code is introduced in detail.According to the different locations of the faulty disk in the disk array,four data recovery algorithms are given when three disks are damaged at the same time in four different cases.The performance of CPC erasure codes is analyzed from three aspects: coding complexity,decoding complexity and update complexity.The scheme is compared with the data layout of existing three-fault-tolerant STAR codes and EEOD(Extending EVENODD,EEOD)codes,and the advantages of the scheme are illustrated with detailed data.3.An energy-efficient data layout scheme called EECPC(energy-efficient cross parity check,EECPC)is proposed.This scheme is an optimization of CRUSH(Controlled Replication Under Scalable Hashing,CRUSH)data layout algorithm,and it can combine with CPC erasure code,which makes up for the limitation of CRUSH algorithm,effectively reduces the energy consumption of RAID storage system and improves the reliability of the system.Through simulation experiments,the EECPC scheme is compared with the STAR scheme,which shows the advantages of EECPC data layout scheme in energy saving.