Indexing

GigaMap supports multiple index types, each optimized for different query patterns.

Index Type Best For Example Query

Index Type	Best For	Example Query
Bitmap	Exact matches, filtering, low/high cardinality	`category = "Electronics"`
Lucene	Full-text search, keyword matching, relevance ranking	`description contains "wireless bluetooth"`
JVector	Vector similarity, semantic search, recommendations	"Find products similar to this one"

Bitmap

Exact matches, filtering, low/high cardinality

category = "Electronics"

Lucene

Full-text search, keyword matching, relevance ranking

description contains "wireless bluetooth"

JVector

Vector similarity, semantic search, recommendations

"Find products similar to this one"

Choosing an Index Type

Bitmap Index - Use for exact value matching and filtering. Ideal for fields like status, category, boolean flags, or any field where you query for specific values. Supports both low cardinality (few unique values) and high cardinality (many unique values) fields.

Lucene Index - Use for full-text search within text content. Ideal for searching product descriptions, article content, user comments, or any text where users search by keywords. Supports relevance ranking, phrase search, wildcards, and boolean queries.

JVector Index - Use for vector similarity search with AI embeddings. Ideal for finding similar items, semantic search, recommendation systems, or RAG applications. Requires vector embeddings from an AI/ML model.