Elassandra documentation

Elassandra embeds OpenSearch inside Cassandra so that each node serves both distributed storage and distributed search.

The maintained line in this repository targets:

• Apache Cassandra 4.0.x

• OpenSearch 1.3.x

• Java 11

The pages in this guide describe the current OpenSearch-based Elassandra implementation, its packaging, and its operating model.

Contents:

• Architecture
  • Concepts Mapping
  • Durability
  • Shards and Replicas
  • Write path
  • Search path
  • Mapping and CQL schema management
• Quick Start
  • Start your cluster
  • Import sample data
  • Create an OpenSearch index from a Cassandra table
  • Create an OpenSearch index from scratch
  • Search for a document
  • Manage indices
  • Cleanup the cluster
  • Docker troubleshooting
• Installation
  • Tarball
  • Deb
  • Rpm
  • Docker image
  • Helm chart
  • Running Cassandra only
• Docker image
  • Build the image locally
  • Start a single node
  • Supported environment variables
  • Filesystem layout
  • Exposed ports
  • Local cluster example
• Helm chart
  • Install with default values
  • Install on minikube
  • Enable OpenSearch Dashboards
  • Provider presets
  • AKS / ACR example
  • Rendered resources
  • Validate before installing
• Configuration
  • Directory Layout
  • Configuration files
  • Logging configuration
  • Multi datacenter configuration
  • Elassandra Settings
  • Sizing and tuning
    • Write performance
    • Search performance
• Mapping
  • Type mapping
  • CQL mapper extensions
  • OpenSearch multi-fields
  • Bi-directional mapping
  • Meta-Fields
  • Mapping change with zero downtime
  • Partitioned Index
    • Virtual index
  • Object and Nested mapping
  • Dynamic mapping of Cassandra Map
    • Dynamic Template with Dynamic Mapping
  • Parent-Child Relationship
  • Indexing Cassandra static columns
  • Elassandra as a JSON-REST Gateway
  • OpenSearch pipeline processors
  • Check Cassandra consistency with OpenSearch
• Operations
  • Indexing
  • GETing
  • Updates
  • Searching
    • Optimizing search requests
    • Caching features
  • Create, delete and rebuild index
  • Open, close index
  • Flush, refresh index
  • Managing Elassandra nodes
  • Backup and restore
    • Restoring a snapshot
    • Point in time recovery
    • Restoring to a different cluster
  • Data migration
    • Migrating from Cassandra to Elassandra
    • Migrating from a legacy search-backed line
  • Tooling
    • JMXMP support
    • Smile decoder
• Search through CQL
  • Configuration
  • Search request through CQL
  • Paging
  • Routing
  • CQL Functions
  • Elassandra aggregations through CQL
  • Application unit tests
  • CQL tracing
• Enterprise
• Integration
  • Integration with an existing Cassandra cluster
    • Rolling replace from Cassandra to Elassandra
    • Create a new Elassandra datacenter
  • Installing OpenSearch plugins
  • Running OpenSearch Dashboards with Elassandra
  • Running Spark with Elassandra
• Testing
  • Testing environment
  • Elassandra build tests
  • Application testing
• Breaking changes and limitations
  • Deleting an index does not delete Cassandra data
  • Nested or object types cannot be empty
  • Document version metadata is not meaningful for global consistency
  • Index and table naming
  • Column names
  • Null values
  • Refresh behavior
  • Unsupported or reduced-scope features
  • Cassandra limitations
• Migrating to Cassandra 4.0.x and OpenSearch 1.3.x
  • HTTP API and clients
  • Dashboards
  • Java runtime
  • Cassandra
  • Implementation status (repository)
  • Target pins (modern stack)
  • Operator validation (staging)
  • Further reading
  • Packaging (when the OpenSearch port ships)
• Cassandra fork inventory (legacy 3.11 line to Apache 4.0.x)
  • Baseline
  • Classification of legacy-fork commits
  • Full ordered list (newest first)
  • Branch design for Cassandra 4.0.x
  • Verification
• Cassandra 4.0.x porting playbook
  • Prerequisites
  • Step 1 — Export patch series (3.11 delta)
  • Step 2 — Bootstrap Apache 4.0 tree
  • Step 3 — Apply patches on 4.0
  • Step 4 — Publish fork and submodule
  • Step 5 — Elassandra JVM integration
• Cassandra 4.0 JVM port (Elassandra integration code)
  • Representative API shifts (non-exhaustive)
  • Primary touchpoints in this repo
• OpenSearch 1.3.x porting guide (codebase map)
  • Recommended approach
    • Clone upstream (side-by-side)
    • Sync org.elassandra.* and rewrite imports (side-car)
  • Bootstrap and process model
  • Cluster state, gateway, discovery
  • Routing and search
  • Metadata and mapping
    • Mapper fork (CqlMapper / CQL columns)
  • Shard coordination
  • Secondary index bridge
  • Modules and tests

Indices and tables

• Index

• Module Index

• Search Page