Network Security Internet Technology Development Database Servers Mobile Phone Android Software Apple Software Computer Software News IT Information

In addition to Weibo, there is also WeChat

Please pay attention

WeChat public account

Shulou

Ceph Learning Notes XIII CephFS file system

2025-01-19 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Servers >

Share

Shulou(Shulou.com)06/03 Report--

CephFS file system

Ceph FileSystem, referred to as CephFS, is a file system storage type that supports POSIX interface. At present, the development lags behind, mainly because the CephFS technology is not mature enough, and because the cloud computing needs more RBD and RADOSGW than CephFS, the community focuses on RBD and RADOSGW. Therefore, at present, CephFS is still more suitable for experiments or test scenarios; with the increase of application requirements, CephFS will be paid more and more attention.

CephFS architecture

The main features of the CephFS file system:

1. The client can be easily mounted and used locally.

2. It can be used for resource sharing.

3. Inherit the fault tolerance and expansibility of RADOS

4. Copy-based redundancy can also be provided to ensure high reliability of data.

CephFS implementation method

Using CephFS requires the use of Metadata Server (MDS) to manage the namespace of the file system and how customers access the OSD data store; MDS is also a service process, so deploy MDS and start the ceph-mds service before experiencing CephFS.

Introduction to MDS (Metadata Server)

What is metadata and its function?

Metadata is mainly responsible for recording the attributes of data, such as file storage location, file size and storage time; responsible for resource search, file record, storage location record, access authorization and so on.

MDS embodiment form

MDS is a service that runs as a process and is mainly responsible for managing files and directories in the CephFS cluster to ensure their consistency. MDS is very similar to MON, so redundancy can be achieved in multi-node deployment.

CephFS mainly relies on MDS processes to provide services. MDS provides a consistent file system with cache layer. MDS does not provide any data directly to the client. All data is provided by the backend OSD, thus reducing the number of reads and writes. MDS also has dynamic clusters, whose MDS can join or exit, and can quickly take over failed nodes. The MDS process can be configured to be active or passive. The active MDS is the master MDS, while the other MDS enters the Standby state. When the master MDS node fails, the Standby node will take over its work and promote the master node; its metadata information responds to external access requests in memory cache.

Processing flow

When the client opens a file, the client sends a request to MDS; MDS locates to OSD the index node where the file is located, which contains the file's metadata information, such as file owner, size, permissions, etc. MDS gives the client the right to read the contents of the cached file and returns to the client File value, Layout (Layout defines how the file content is mapped to Object), and file size when access is authorized According to the information returned by MDS, the client locates to the file that needs to be accessed, and then interacts directly with OSD.

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.

Views: 0

*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.

Share To

Servers

Wechat

© 2024 shulou.com SLNews company. All rights reserved.

12
Report