A File System Using GPU-Accelerated File-wise Reliability Scheme
Chien-Kai Tseng, Shang-Chieh Lin, Yarsun Hsu · 2012
This work revises the original file-wise reliability scheme to cope with larger pages in storage devices nowadays, and implements it as a file system prototype: CRSFS. There are four layers in CRSFS: GPU primitive for Cauchy Reed-Solomon (CRS) coding, CrystalGPU framework, CRS coding layer and AFS FUSE layer. CRSFS provides GPU acceleration on the CRS encoding/ decoding operations by using the CUDA program: GPU primitive for CRS coding. Besides, it is integrated with FUSE (Filesystem in Userspace) framework and Rx (extended remote procedure call) protocol in AFS FUSE layer to provide high flexibility on storage system configurations. Hence, most programs can benefit from it without rewriting their read/