🗒️ ReductStore Glossary
bucket
A container for storing data in ReductStore. Buckets are uniquely named and can have configurable storage settings, such as quotas and replication policies. Buckets are used to separate data by source or purpose and support FIFO or HARD quotas for managing storage efficiently.
Related entries: entry, record, replication, FIFO quota, HARD quota
entry
A uniquely named source of data within a bucket. Entries help organize and separate data sources or purposes. Each entry contains records, which are stored in batches for performance optimization.
Related entries: bucket, record, block
record
A piece of data stored in an entry, consisting of metadata (timestamp, content type, labels) and content (actual data). Records are the fundamental data units in ReductStore and are stored within blocks.
block
A batch of records stored on disk. Blocks store records in fixed-size segments, optimizing data access by time ranges. Each block includes a descriptor file (metadata) and a data block (actual data).
Related entries: record, entry, batching
batching
A method of writing or reading multiple records at once to improve performance. Batching reduces HTTP overhead and optimizes data ingestion and retrieval in ReductStore. The storage engine stores records in blocks to minimize disk space usage.
Related entries: block, record
FIFO quota
A "First-In-First-Out" quota based on the volume of stored data in a bucket. When the storage volume in a bucket reaches its specified limit, the oldest blocks are removed automatically to make room for new data. This ensures efficient use of storage without manual intervention.
Related entries: bucket, block
HARD Quota
A strict limit on the total storage capacity allocated to a bucket. When the storage usage reaches the specified limit, no new records can be added until space is freed by deleting existing data. This ensures complete control over storage usage within predefined limits.
Related entries: bucket, block
SDK
Software Development Kits provided by ReductStore for popular programming languages. SDKs simplify interaction with the HTTP API by abstracting the need to write HTTP requests manually.
Related entries: HTTP API
HTTP API
The primary interface for interacting with ReductStore. It enables clients to write, read, filter, and manage data using HTTP requests. The API supports batching and label-based conditional querying.
Related entries: batching, conditional query
conditional query
A query language based on JSON syntax used to filter records in ReductStore using labels and their values. It supports multiple conditions, nested queries, and two primary notations: object notation and array notation.
Related entries: HTTP API, record, replication
Web Console
A graphical interface for managing ReductStore instances. The web console allows users to configure buckets, monitor data ingestion, and manage replication settings.
Related entries: bucket, replication
CLI
The Command Line Interface for ReductStore. It provides a way to interact with the storage engine through terminal commands, enabling tasks such as copying records, configuring buckets, and managing replication.
Related entries: HTTP API, replication
replication
The process of copying data from a source bucket to a destination bucket. ReductStore supports append-only replication, allowing filtered records to be replicated while maintaining write-ahead logs for reliability.
Related entries: bucket, conditional query
access token
A token-based authentication and authorization method in ReductStore. Access tokens ensure secure access to storage, with options to enable or disable authentication as required.
Related entries: HTTP API