Filesystem

In the last few years, the need for compact and embedded systems has expanded in all fields. With regard to this development, ARM development platform is the ideal and practical answer for planning a new product design. ARM platforms carry a generally positive result regarding speed, accuracy, adaptability, size and cost. Every new embedded design, at the back an operating system must be modified appropriately for the specifications of the embedded target. It would be a huge added advantage to have an embedded design running an OS. In particular, this paper underlines the BSP (Board Support Package) customization of Linux operating system and porting mechanism to FreeScale SabreSD an i.MX6 processor based embedded board with the help of Yocto build environment. The usual method of going for Linux Target Image Builder (LTIB) is considered old and Yocto is specifically preferred as Freescale is emphasizing on lighter and easier version. Yocto is a dream project by Freescale. Successful build of Yocto environment enables customization of kernel, through which images of OS are built and they are ported to target platform.

Sommario Questo documento descrive i cluster file system: nel primo capitolo vengono illustrate le motivazioni che hanno portato al loro sviluppo ed utilizzo nel settore dell'High Performance Computing (HPC), partendo da un'introduzione alle varie tipologie di file system. Viene, inoltre, descritta nel dettaglio la terminologia utilizzata per descrivere le funzionalità di questa tipologia di file system.

Conventional file systems are optimized for large file transfers instead of workloads that are domi-
nated by metadata and small file accesses. MetaCache stores file system metadata into LSM Tree. The LSM-tree is a hybrid (two tree-like components) data structure.We are creating cache just below VFS which stores the metadata of files .Our project shows clear enhancement in metadata handling which dominate the performance of all otherlocal file systems (xfs,ext4 etc.)

Nowadays, Linux file systems have to manage millions of tiny files for different applications, and face with higher metadata operations. So how to provide such high metadata performance with such enormous number of files and large scale directories is a big challenge for Linux file system. We viewed that metadata lookup operations dominate metadata workload and incur low metadata performance. In this paper, we present a metadata cache to accelerate metadata access for Linux file system. Through this optimization, the Linux file system (such as EXT2, EXT4, BTRFS, etc.) can gain improvement in read rates as well as write rates.