定义 A file system (often also written as filesystem ) is a method of storing and organizing computer files and their data. Essentially, it organizes these files into a database for the storage, organization, manipulation, and retrieval by the computer's operating system.
How to scale本来应该说是 Scale Your Storage 这个 topic 太大 涉及到应用,硬件,网络 先简单的谈分布式文件系统 谈点穷人的方案
9.
看看我们有的文件系统 太多了…看分类吧 Diskfile systems –ext3/ntfs/zfs/wafl ( 大部分我们熟悉的 ) File systems with built in fault-tolerance-zfs/brfs File systems optimized for flash memory, solid state media Record-oriented file systems Shared disk file systems Distributed file systems Distributed fault-tolerant file systems Distributed parallel file systems Distributed parallel fault-tolerant file systems GoogleFilesystem/CloudStore/Lustre/HDFS Peer-to-peer file systems Special purpose file systems Pseudo- and virtual file systems Encrypted file systems
10.
Google File systemLike Google File system KFS HDFS Why like? Master – chunk 架构 POSIX Like Interface 设计目标一致
11.
GFS Goal 最开始是为爬虫等应用设计的The system is built from many inexpensive commodity components that often fail. The system stores a modest number of large Files. large streaming reads and small random reads The workloads also have many large, sequential writes that append data to Files High sustained bandwidth is more important than low latency
MogileFS Application levelNo single point of failure Automatic file replication "Better than RAID" Flat Namespace Shared-Nothing / No RAID required Local filesystem agnostic
FastDFS 构成 TrackerServer 主要做调度工作,在访问上起负载均衡的作用。记录 storage server 的状态,是连接 Client 和 Storage server 的枢纽。 Storage Server 存储服务器,文件物理内容和 meta data 都保存到存储服务器上 Storage Server 构成不同的组 ( 卷 /Volume) 同组的 Storage Server 的文件是相同的
#10 SSD - 嵌入式系统 , 写入优化 Record – 是记录方式的,区别于大部分文件系统的 Shard-disk SAN- Redhat GFS 等 DFS - SMB is also known as Common Internet File System (CIFS) DFS-FT MS DFS / DFS-Parallel 没啥好说的,还有下面 DFS-Parallel-FT GoogleFilesystem / CloudStore / Lustre [Laster]
#31 Namespace 的查找是一个消耗很大的操作 , NFS 文件句柄缓存 NFS 文件句柄缓存 Facebook have extended the Linux kernel to allow NFS file opens via inode number rather than filename to avoid the NetApp scaling issue. Namespace 可以扁平化