Write Performance Analysis with Buffered FUSE
نویسندگان
چکیده
With the development of technology, the size of files handled by smart devices is increasing. This phenomenon can cause the storage to be the biggest bottleneck of smart devices. There is a study that presents a major factor in this phenomenon is overhead in user level filesystem (FUSE). And it proposed buffered FUSE (bFUSE) and increased size of FUSE I/O as a solution of this problem. In this paper, we reproduced it in ODROID-Q2 and measured write performance. As a result, we can raise the write performance up to 96% compared to raw device bandwidth with FUSE.
منابع مشابه
Buffered FUSE: optimising the Android IO stack for user-level filesystem
In this work, we optimise the Android IO stack for user-level filesystem. Android imposes user-level filesystem over native filesystem partition to provide flexibility in managing the internal storage space and to maintain host compatibility. The overhead of user-level filesystem is prohibitively large and the native storage bandwidth is significantly under-utilised. We overhauled the FUSE laye...
متن کاملAN EFFICIENT METHOD FOR OPTIMUM PERFORMANCE-BASED SEISMIC DESIGN OF FUSED BUILDING STRUCTURES
A dual structural fused system consists of replaceable ductile elements (fuses) that sustain major seismic damage and leave the primary structure (PS) virtually undamaged. The seismic performance of a fused structural system is determined by the combined behavior of the individual PS and fuse components. In order to design a feasible and economic structural fuse concept, we need a procedure to ...
متن کاملAdaptation of Distributed File System to VDI Storage by Client-Side Cache
Distributed File system was widely used as a cloud computing backend storage with high scalability, low cost, and reasonable sequential I/O performance. VDI I/O workload is mostly composed of small random I/O and distributed file system does not have enough performance for this I/O pattern. Compensating for this gap, we applied a cache using memory and SSD (Solid State Drive) to distributed fil...
متن کاملAndroStep: Android Storage Performance Analysis Tool
Abstract: The applications in Android based smartphones generate unique IO requests; however, existing IO workload generators and trace capturing tools are designed to neither generate nor capture the IO requests of Android apps. In this paper, we introduce the Android storage performance analysis Tool (AndroStep) which is specifically designed for characterizing and analyzing the behavior of t...
متن کاملTrail: A Fast Synchronous Write Disk Subsystem Using Track-Based Logging
Synchronous disk write presents a serious performance problem for those applications that require strong data integrity guarantee. This paper describes the design and implementation of Trail, a disk subsystem that features a track-based logging technique to minimize the performance impact of synchronous disk write operations. We have demonstrated in this research that Trailcan reduce the latenc...
متن کامل