Glossary
Understanding MooseFS requires familiarity with key terms and concepts related to distributed storage. This glossary provides clear definitions of important terminology used throughout the documentation, helping users navigate technical details with ease.
Whether you're new to MooseFS or need a quick reference, this section will clarify essential terms and ensure a better grasp of the system's functionality.
Master Server
The central component of MooseFS responsible for managing metadata, coordinating file operations, and ensuring the overall consistency of the file system. It keeps track of file locations, access permissions, and data replication but does not store actual file content. A high-availability setup (only in PRO version) can include backup masters for failover support.
Leader Master Server (PRO)
The active Master Server in a high-availability MooseFS setup. It manages metadata, coordinates file operations, and communicates with clients and chunkservers. The Leader Master Server is responsible for making real-time decisions regarding file system structure, access control, and data distribution. If it fails, a Follower Master Server can take over to ensure system continuity.
Follower Master Server (PRO)
A standby Master Server that continuously synchronizes metadata from the Leader Master Server. It remains in a passive state, ready to take over if the Leader fails. This redundancy mechanism enhances system reliability and minimizes downtime in high-availability configurations.
Metalogger
A supporting service that periodically retrieves and stores metadata snapshots from the Master Server. This component enhances system resilience by providing a backup of critical metadata, which can be used for recovery in case of a Master Server failure.
Chunkserver
A storage node that holds actual file data in the form of chunks. Chunkservers communicate with the Master Server, reporting available storage and responding to read and write requests. Data redundancy is achieved by replicating chunks across multiple chunkservers to ensure fault tolerance and reliability.
Redundancy Level
The number of copies (replicas) of each data chunk stored across multiple chunkservers to ensure fault tolerance and data availability. A higher redundancy level provides greater protection against hardware failures but requires more storage space. MooseFS dynamically manages redundancy to maintain the required number of replicas.
Goal
A file-specific setting in MooseFS that defines the desired redundancy level. The goal determines how many copies of a file's chunks should exist in the system. Users can configure different goals for different files or directories, balancing storage efficiency and reliability based on specific needs.