Startups > Firebolt

Firebolt

Firebolt enables sub-second SQL queries on petabyte-scale data at high concurrency.

Series C $269M total Founded 2019 Tel Aviv, Tel Aviv

Firebolt is a cloud data warehouse built for sub-second query performance on massive datasets at high concurrency. It decouples metadata, storage, and compute to enable independent scaling, allowing organizations to run complex SQL queries on petabyte-scale data with millisecond latency. The platform supports diverse workloads from batch analytics to customer-facing AI applications, with real-world customers like Similarweb querying petabytes in production while ingesting terabytes daily.

Problem solved

Organizations struggle to run complex, low-latency SQL queries on massive datasets while maintaining high concurrency without prohibitive infrastructure costs.

Target customer

Enterprise data teams and analytics organizations processing large-scale datasets with demanding latency and concurrency requirements; particularly suited for companies building customer-facing data applications and AI workloads

Website LinkedIn Crunchbase

Founders

Eldad Farkash

CEO

Co-founder of business intelligence unicorn Sisense; led Firebolt from inception.

Saar Bitner

COO

Previously COO at Sisense; co-founding leadership at Firebolt.

Ariel Yaroshevich

Co-Founder

Sisense alum; initially served as CTO before Mosha Pasumansky joined from Google BigQuery.

Funding history

Series A $37M December 2020 Led by Zeev Ventures · TLV Partners, Bessemer Venture Partners, Angular Ventures

Series B $127M June 2021 Led by Dawn Capital · K5 Global, Zeev Ventures, TLV Partners, Bessemer Venture Partners, Angular Ventures

Series C $100M January 2022 Led by Alkeon Capital · Sozo Ventures, Glynn Capital, Zeev Ventures, Angular Ventures, Dawn Capital, Bessemer Venture Partners, K5 Global, TLV Partners

Total raised: $269M

Industries

Cloud Data Services Big Data Analytics Warehousing SaaS

Pricing

Usage-based model with pay-as-you-go or committed-use discount options. Compute billed per-second based on node type (FBUs), plus separate data storage costs. Choice of compute-optimized or storage-optimized engines. Flexible for startups; discounted rates available for consistent, high-volume workloads.

Notable customers

Similarweb, IQVIA, WingX, Lurkit, Ezora

Website

firebolt.io/

Competitors

Snowflake

Firebolt delivers 8x better price-performance and 5.5x higher query throughput; Snowflake costs 37.5x more to achieve equivalent Firebolt performance.

Amazon Redshift

Firebolt is 18x better in price-performance and delivers 10x higher query throughput; Redshift failed to match Firebolt's performance in benchmarks.

Google BigQuery

Firebolt is 90x better in price-performance; BigQuery cannot achieve Firebolt's sub-second latency and high-concurrency characteristics at comparable cost.

Why this matters: Firebolt represents a significant architectural rethink of cloud data warehousing, decoupling metadata, storage, and compute to deliver unprecedented sub-second performance and concurrency at a fraction of competitor costs. The company's pedigree (founders from Sisense unicorn, CTO from Google BigQuery) and substantial Series C funding ($269M total) position it as a credible challenger to entrenched players like Snowflake and BigQuery.

Best for: Enterprise organizations needing to query massive datasets with sub-second latency and high concurrency for analytics, AI applications, and customer-facing data services.

Use cases

High-Concurrency Analytics at Petabyte Scale

Similarweb uses Firebolt to maintain a large-scale analytics platform serving thousands of customers with sub-second performance while ingesting five terabytes daily. The decoupled architecture enables isolation of concurrent workloads without performance degradation.

Patient Record Analytics for Healthcare

IQVIA queries over 1 billion patient records in milliseconds across diverse workloads from ELT to batch analytics to customer-facing applications. Firebolt's performance enables real-time insights on massive healthcare datasets.

Cost-Optimized Historical Data Analysis

Lurkit serves data from millions of gaming channels to 50k customers while reducing analytics costs by 40% and enabling 10x larger historical queries. The platform eliminates expensive aggregation requirements through Firebolt's sub-second performance.

Real-Time Market Intelligence at Scale

WingX scans 168 billion rows daily to provide aviation companies critical market insights, achieving 70% faster query performance compared to traditional SQL Server warehouses.

Alternatives

Snowflake Choose Snowflake if you prioritize ecosystem breadth and mature feature set over price-performance; Snowflake is more expensive but has broader market adoption.

Amazon Redshift Choose Redshift if you're deeply invested in AWS and need tight integration; Redshift cannot match Firebolt's sub-second latency and concurrency at comparable cost.

Google BigQuery Choose BigQuery if you prioritize Google Cloud ecosystem integration; BigQuery is significantly more expensive to achieve Firebolt's performance levels.

FAQ

What does Firebolt do? +

Firebolt is a cloud data warehouse optimized for sub-second SQL query performance on massive datasets at high concurrency. It decouples storage and compute for independent scaling, enabling organizations to run complex analytics and customer-facing data applications on petabyte-scale data with millisecond latency.

How much does Firebolt cost? +

Firebolt uses a usage-based pricing model with per-second compute billing based on node type (measured in FBUs) plus data storage costs. Two options are available: pay-as-you-go for flexible, unpredictable workloads, or committed-use discounts for consistent, high-volume usage. Contact sales for enterprise pricing.

What are alternatives to Firebolt? +

Top alternatives include Snowflake (broader ecosystem but higher cost), Amazon Redshift (AWS-native but lower performance), and Google BigQuery (Google Cloud integration but significantly more expensive for Firebolt-equivalent performance).

Who uses Firebolt? +

Notable customers include Similarweb, IQVIA, WingX, Lurkit, and Ezora. Target customers are enterprise data teams processing large-scale datasets with demanding latency and concurrency requirements, particularly those building customer-facing analytics and AI applications.

How does Firebolt compare to Snowflake? +

Firebolt delivers 8x better price-performance than Snowflake and 5.5x higher query throughput. To achieve Firebolt's performance characteristics, Snowflake costs 37.5x more. Firebolt excels at sub-second latency and high-concurrency workloads, while Snowflake offers broader ecosystem maturity.