Shingled magnetic recording (SMR) is regarded as a promising solution for fulfilling the capacity requirement of next-generation big data applications. However, due to the sequential-write constraint of SMR drives, random-write requests could only be achieved via read-merge-write (RMW) operations, which leads to a huge amount of write amplification. SMR drives nowadays employ an on-disk persistent cache area for absorbing random-write traffic. Nevertheless, the on-disk persistent cache area cannot absorb duplicate write traffic via overwriting due to the sequential-write constraint. Such observation motivates us to propose a sequential-write-constrained cache (SWC2) management to mitigate the write amplification issue of SMR drives. The proposed strategy is implemented within the Linux system and the experimental results show that the amount of write amplification can be reduced by an average of 52.27%.
SAC '19 Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing, p.599-606