DataTerminology.comNEW
Also known as: PrestoSQL
An open-source distributed SQL engine that federates fast analytics across many data sources at scale.
DESCRIPTION
Trino is a high-performance MPP SQL engine forked from Presto, focused on rapid development and broad connector support. It enables interactive queries across object stores, Hive metastore, relational databases, NoSQL systems, and streaming platforms. Trino emphasizes query federation, predicate pushdown, and efficient execution with memory and CPU optimizations. Its coordinator-workers architecture supports elastic scaling and resource groups. Use cases include BI dashboards, exploration, and lakehouse querying. Trino often integrates with catalogs like Hive or Glue and formats like Parquet and ORC. Ecosystem vendors, such as Starburst, provide enterprise features. It is frequently compared with Presto and Spark SQL.
ALTERNATIVE DEFINITIONS
Trino Project
Trino is a distributed SQL query engine designed to query large data sets from multiple data sources.
Stresses multi-source federation and scalability.
AWS EMR Documentation
Trino is a fast distributed SQL query engine for big data analytics that supports querying data from various sources.
Emphasizes speed and multi-source capabilities on EMR.
Starburst
Trino is an open source distributed SQL query engine for data analytics.
Positions Trino as open source foundation for enterprise analytics.
RELATED TERMS
Starburst Enterprise Apache Hive Query Federation Middleware ANSI SQL Data Lake Data Lakehouse Apache Kafka
COMPETING TERMS
RELATIONS
- Conforms to: ANSI SQL
KNOWLEDGE AREAS
Data Engineering Data Warehousing & BI Platform & Tooling
TARGET AUDIENCE
Data Engineer Analytics Engineer BI Specialist Data Scientist Data Architect
EXTERNAL SOURCES
- Wikipedia - Trino
- Trino Project - Official Trino documentation
- CNCF - Community overview of Trino
TOP QUESTIONS
Why choose Trino for analytics on a data lake?
Trino delivers fast SQL over lake formats with federation, reducing data movement and enabling interactive analysis.
Category: Relevancy
How does Trino differ from Presto?
Trino, a community fork, progresses faster with wider connectors and features; both share architecture and goals.
Category: Comparison
What are Trino’s typical deployment patterns?
Run a coordinator with autoscaled workers on Kubernetes or cloud instances; configure catalogs for each data source.
Category: Management
What mistakes slow down Trino queries?
Ignoring partitioning, using inefficient file formats, and missing predicate pushdown can degrade performance significantly.
Category: Mistakes
How do you secure Trino in production?
Use TLS, authentication, fine-grained access via catalogs, network isolation, and audit logging.
Category: Requirements
Which file formats work best with Trino?
Columnar formats like Parquet and ORC provide compression and efficient predicate pushdown.
Category: Best practices
What hot topics involve Trino now?
Cost-based optimization, lakehouse table format support, and autoscaling strategies are key areas.
Category: Topic
Can Trino be used for ETL writes?
It’s primarily read-optimized; writes exist but are limited. Use Spark or ELT tools for heavy writes.
Category: Misunderstanding
What risks come with Trino’s federation?
Complex cross-source joins can be costly; governance and query limits prevent resource contention.
Category: Risks
Where should beginners start learning Trino?
Begin with the official docs, connectors guide, and a small Kubernetes or Docker setup.
Category: Getting started