Firebolt was architected from the ground up to leverage the benefits of the cloud and modern data stack, so you can build robust & scalable analytics experiences that will last.
A simple Saas platform
Focus on what matters
Enjoy the simplicity of SaaS and get started in minutes. Let Firebolt worry behind the scenes about things like provisioning, scale, and storage optimization, so that you can focus on generating value and insight from your data.
Built for
Native decoupling of storage & compute
Architectures that still couple storage & compute are just not acceptable anymore for modern data challenges. Decoupling allows you to match any workload with the right compute resources in complete isolation, easily scale up/down, and be in control of price/performance tradeoffs.
optimize price-performance
Granular control over compute resources
At Firebolt we believe that modern data engineering & dev teams value the ability to control what they’re running. This is why Firebolt lets you be very granular in your compute resource choices, letting you choose different CPU, SSD, and RAM combinations. This will allow you to optimize your resources for the best price-performance ratio, without over-spending on under-utilized resources.
Effortless integration
Leverage your data lake on S3
Whether you store data in Parquet, ORC, CSV or JSON - just point Firebolt to S3 and easily set up continuous ingestion pipelines. Quickly get up and running with your most challenging workloads in Firebolt, while keeping what works in place.
A look inside
The firebolt architecture
SQL Firebolt is ANSI SQL compliant, and lets you manage all aspects of the platform through SQL. Use a variety of clients to use and programmatically orchestrate everything about Firebolt.
SaaS Firebolt is fully SaaS. Our multi-tenant services layer handles all requests and all security, administration, and metadata aspects.
Firebolt Engines Firebolt engines are compute clusters that can be granularly configured for optimal price-performance, and run decoupled from storage in complete isolation. Scaling is a click away for easy elasticity. All clusters run in isolation per client.
Copy data from S3 Point a Firebolt external table to your data in S3 to create a pipeline that ingests it continuously into Firebolt.
Storage layer Data that is ingested into Firebolt is persisted to S3 in Firebolt’s sorted, compressed and indexed F3 format. Indexes are persisted to S3 as well. Data is pulled from your VPC and into Firebolt’s SaaS, where it is stored in isolation.
