Shulou(Shulou.com)06/02 Report--

GlusterFS Overview GlusterFS Introduction Open source distributed file system consisting of storage servers, clients, and NFS/Samba storage gateways Features of GlusterFS No metadata server Scalability and high performance High availability Globally unified namespace Elastic volume management based on standard protocols GlusterFS terminology Brick: storage node Volume: volume FUSE: kernel module, interaction module on the client VFS: virtual port Glusterd: Services Modular Stack Architecture Modular, Stacked Architecture Responsible functionality is achieved through the combination of modules GlusterFS work model GlusterFS workflow

Application： Client or application accesses data via GlusterFSync mount point VFS: Linux kernel receives request via VFS API and processes FUSE: VFS delivers data to FUSE kernel file system, fuse file system delivers data to GlusterFS client via/dev/fuse device file GlusterFS Client ; The elastic HASH algorithm obtains a 32-bit integer divided into N consecutive subspaces through the HASH algorithm, and each space corresponds to a Brick. The advantages of the elastic HASH algorithm ensure that the data is evenly distributed in each Brick. Further solve the single point of failure and access bottleneck GlusterFS volume type distributed volume does not partition files through extended file attributes HASH value support for underlying file systems Ext3, ext4, ZFS, XFS and other characteristics of distributed volume files distributed on different servers. No redundancy Easier and cheaper to expand the size of a volume Single point of failure causes data loss Create a distributed volume based on the underlying data protection Create a distributed volume named dis-volume, files will be distributed according to HASH among server1:/dir1, server2:/dir2, and server3:/dir3

gluster volume create dis-volume server1:/dir1 server2:/dir2 Stripe volume divides files into N blocks (N stripe nodes) based on offset, polled storagePerformance is particularly good when storing large files on each Brick Server nodeNo redundancy, Raid0 like featuresData is divided into smaller blocksDistribution of different stripes in a block server farm reduces load and smaller filesSpeed of accessNo data redundanceCreate a stripe volume called Stripe-volume, files will be chunked and polled for storage in two bricks, Server1:/dir1 and Server2:dir2

gluster volume create stripe-volume stripe 2 transport tcp server1:/dir1 server2:/dir2 replication volume keeps one or more copies of the same file Replication mode Because copies are kept, disk utilization is low. Storage space on multiple nodes is inconsistent, so the capacity of the lowest node is taken as the total capacity of the volume according to barrel effect. All servers in the volume keep a complete copy. The number of copies of the volume can be determined by the customer when creating it. At least two block servers or more servers are redundant. Create a replication volume named rep-volume. The file will store two copies at the same time.

gluster volume create rep-volume replica 2 transport tcp server1:/dir1 server2:/dir2 distributed stripe volumeThe function of both distributed and stripe volumes is mainly used for large file access processingAt least 4 servers are required to create a distributed stripe volumeCreate a distributed stripe volume named dis-stripe. When configuring a distributed stripe volume, the number of storage servers contained in the Brink must be a multiple of the number of stripes (>=2 times)

luster volume create rep-volume stripe 2 transport tcp server1:/dir1 server2:/dir2 server3:/dir1 server4:/dir2 distributed replication volumeFunctions of both distributed and replicated volumes are used to create distributed replication volumes when redundancy is required Create a distributed stripe volume named dis-rep, when configuring replication volumes for distributed stripe volumes, the Brink in the volume must contain a multiple of the number of stripes (>= 2x)

gluster volume create rep-volume replica 2 transport tcp server1:/dir1 server2:/dir2 server3:/dir1 server4:/dir2

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.