Firebolt continuously releases updates so that you can benefit from the latest and most stable service. These updates might happen daily, but we aggregate release notes to cover a longer time period for easier reference. The most recent release notes from the latest version are below.

Firebolt might roll out releases in phases. New features and changes may not yet be available to all accounts on the release date shown.

​
Firebolt Release Notes - Version 4.24

​
New Features

Introduced Organization-Level Role-Based Access Control (ORBAC)

Firebolt has introduced Organizational-Level Role-Based Access Control (ORBAC), extending traditional account-level RBAC to provide centralized control over identities and permissions across all accounts in an organization. ORBAC enables administrators to manage global resources such as accounts, logins, service accounts, network policies, and organization roles from a unified model.

Key highlights include:

  • Global Scope: ORBAC operates across the entire organization rather than being limited to individual accounts.
  • Strict Deny-by-Default: No access is granted by default; all permissions must be explicitly assigned via Organization Roles.
  • New System Role: The built-in org_admin role is granted all organization-level privileges by default.
  • New Information Schema Views: Added views like enabled_roles, applicable_roles, transitive_applicable_roles, and object_privileges under org_db.information_schema for visibility into role assignments and access.

Example usage:

-- Create a new organization role
CREATE ORGANIZATION ROLE account_manager_role;

-- Grant permission to manage all accounts
GRANT MODIFY ANY ACCOUNT TO ORGANIZATION ROLE account_manager_role;

-- Assign role to a login
GRANT ORGANIZATION ROLE account_manager_role TO LOGIN "john@acme.com";

ORBAC provides enterprise-grade governance for managing organizational access, complementing existing account-level RBAC. It is ideal for administrators managing multiple Firebolt accounts and seeking consistent, scalable permission models.

Added support for PostgreSQL INSERT ON CONFLICT syntax The optional ON CONFLICT extension to the INSERT statement provides a way to reconcile or merge new data with existing records by either skipping or overwriting rows that already exist. It is available for a narrow subset of INSERT statements. For more generic UPSERT or deduplication functionality, use the MERGE statement.

Example usage:

-- Insert and update if conflict occurs on column 'did'
INSERT INTO distributors (did, dname) VALUES (5, 'Gizmo Transglobal')
    ON CONFLICT (did) DO UPDATE SET dname = EXCLUDED.dname;

-- Insert and ignore if conflict occurs on column 'did'
INSERT INTO distributors (did, dname) VALUES (7, 'Redline GmbH')
    ON CONFLICT (did) DO NOTHING;

Documentation is available here.

Added background warmup feature to engines

Added a feature that allows engines to proactively fetch recently used data at startup. While engines are running, they will now periodically record recently read files, and when engines start, they will proactively fetch these files from s3.

This feature can be enabled / disabled in SQL. For example:

-- Enable auto_warmup on myengine
ALTER ENGINE myengine SET AUTO_WARMUP = true;

-- Disable auto_warmup on myengine
ALTER ENGINE myengine SET AUTO_WARMUP = false;

Added release channel configuration to engines

Added a feature that allows to configure per-engine release settings, in particular, to specify release a channel that controls whether the engine gets upgraded normally or can get an earlier preview version. Available values are: DEFAULT or PREVIEW to explicitly allow early upgrades.

Examples:

  1. Create a new engine and assign it to the preview channel:
CREATE ENGINE myengine WITH RELEASE = (CHANNEL = PREVIEW);
  1. Update the engine to assign it back to the default channel:
ALTER ENGINE myengine SET RELEASE TO (CHANNEL = DEFAULT);

Updated ALTER ENGINE ... SET syntax

Syntax of configuration parameters assignment is aligned with PostgreSQL and allows to assign values using TO keyword as an equivalent of equal sign.

For example to change auto stop configuration:

ALTER ENGINE myengine SET AUTO_STOP TO 30;

Introduced new ingestion related setting

Introduced max_insert_threads to control the maximum number of threads for INSERT statements, limiting the degree of parallelism for tablet writing operations. This can reduce memory footprint during ingestion.

Output of EXPLAIN

  • The output for accessing an aggregating index via TABLESCAN is now represented as ”@” instead of “__AGG_IDX_TABLE”.
  • Included GRANULES in EXPLAIN ANALYZE output. This provides users with deeper insights into query performance and helps optimize indexing strategies.

Added telemetry data to the telemetry column in information_schema.engine_query_history The telemetry column in information_schema.engine_query_history now shows telemetry data about query execution. This enhancement helps users to better analyze the performance of their queries.

Added the peak_memory_bytes column to the information_schema.engine_query_history table to monitor query memory usage The information_schema.engine_query_history table now contains the peak_memory_bytes column, which provides details on query memory usage. This addition helps users monitor and optimize query performance.

​
Performance Improvements

Enhanced the VACUUM statement to merge tablets from different nodes into a single tablet

The VACUUM statement now merges tablets from different nodes into a single tablet. This enhancement improves data management efficiency and optimizes storage space.

Enabled filter expressions to be pushed into PREWHERE above joins with configuration for primary index column usage in PREWHERE Firebolt now allows filter expressions to be pushed down into PREWHERE above joins. By default, expressions on the first primary index column are not chosen for PREWHERE, as they are used for granule pruning.

A new setting, ALLOW_PREWHERE_ON_FIRST_PRIMARY_INDEX_COL, lets the optimizer select predicates on the first primary index column for PREWHERE. This change can improve performance when the first primary index column has high cardinality and the query accesses additional wide columns.

​
Connectors

Added support for Pandas

Firebolt now supports Pandas, allowing to interact with Firebolt data using Pandas DataFrames. A guide on how to use Pandas with Firebolt is available here.

Added support for Tableau Prep

Firebolt now supports Tableau Prep, allowing to prepare and clean data before analysis. A guide on how to use Tableau Prep with Firebolt is available here.