Zhu Yuan, Ping Xie, Xiaojuan Zhang, Shengling Geng, Jun Ma
{"title":"Accelerate RAID scaling by reducing disk I/Os and XOR operations","authors":"Zhu Yuan, Ping Xie, Xiaojuan Zhang, Shengling Geng, Jun Ma","doi":"10.1145/3318265.3318266","DOIUrl":null,"url":null,"abstract":"In order to suffice the storage requirements under the big data environment, scaling method is generally adopted to increase the storage capacity of the storage system with the exponential growth of data in the current. RAID has received wide attention in the academic and the industry due to good independence and redundancy. After long-term development, RAID has derived different RAID levels according to distinct requirements of users. In order to achieve load balance including old and new, some data need to be migrated from old data disks to new data disks. However, there will generate the disk I/O operations and the XOR computational operations because of the existence of independent parity disk during the migration process. These affect the efficiency of scaling to a certain extent and lead the problem of long scaling time. In this paper, the scaling process for RAID-4 is optimized by reducing the disk I/O operations and the XOR computational overhead. In the comparison experiment analysis of real storage system scaling, the scaling time of optimization approach is reduced by 49.9% to 57.3% compared with the traditional scaling approach.","PeriodicalId":241692,"journal":{"name":"Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications","volume":"8 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3318265.3318266","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In order to suffice the storage requirements under the big data environment, scaling method is generally adopted to increase the storage capacity of the storage system with the exponential growth of data in the current. RAID has received wide attention in the academic and the industry due to good independence and redundancy. After long-term development, RAID has derived different RAID levels according to distinct requirements of users. In order to achieve load balance including old and new, some data need to be migrated from old data disks to new data disks. However, there will generate the disk I/O operations and the XOR computational operations because of the existence of independent parity disk during the migration process. These affect the efficiency of scaling to a certain extent and lead the problem of long scaling time. In this paper, the scaling process for RAID-4 is optimized by reducing the disk I/O operations and the XOR computational overhead. In the comparison experiment analysis of real storage system scaling, the scaling time of optimization approach is reduced by 49.9% to 57.3% compared with the traditional scaling approach.