Astronomer, the company behind Astro, the leading data orchestration and observability platform powered by Apache Airflow®, today announced the findings from its State of Airflow Report 2025. Based on responses from over 5,000 data practitioners – making it the largest data engineering survey to date – the report reveals the growing criticality of data orchestration in enterprise operations: from advanced analytics to revenue-generating, external-facing data products to production AI and MLOps.
Also Read: AI and Big Data Governance: Challenges and Top Benefits
Apache Airflow has firmly established itself as the industry standard for data orchestration, a position reinforced by its remarkable growth and widespread adoption. The State of Airflow Report 2025 clearly shows a surge in Airflow popularity, usage and community growth by taking a look at metrics from November of last year and the same time period in the year 2020, when the last major release–Airflow 2.0–came out. As of November 2024, Airflow saw:
- Tremendous increase in popularity: Airflow had more than 31M+ downloads in November 2024, compared to 888K+ in November 2020. In fact, there were more Airflow downloads in 2024 than all previous years combined.
- A growing contributor community: Airflow has 3,000+ Airflow contributors, compared to 1,300+ in November 2020. Airflow has the most contributors among all Apache Software Foundation open source projects, including others like Apache Kafka® and Apache Spark®.
- Increasing pull request volume: 29K+ pull requests in November 2024, compared to 10K+ in November 2020
- Exponential enterprise adoption: 77K+ organizations were using Airflow as of November 2024, compared to 25K+ in 2020.
This remarkable growth underscores both Airflow’s increasing popularity among data professionals as well as its continual expansion of use, as organizations in every industry turn to Airflow to support more business functions.
“The State of Airflow Report 2025 confirms what we’re seeing across industries – leveraging data effectively is becoming a competitive necessity,” said Andy Byron, CEO, Astronomer. “Data orchestration has moved from a business need to a strategic imperative, serving as the backbone of modern enterprise infrastructure. With over 90% of data professionals citing Airflow as critical to their business, it’s clear that orchestration is foundational for operationalizing data–fueling everything from advanced analytics to AI-driven innovation–and that Airflow is the standard for data orchestration.”
Notable themes that emerged from the report include:
- Data platform diversity is the new multi-cloud – While market analysts may be looking for a clear winner in the cloud data platforms battle, enterprises have clearly chosen a multi-solution strategy–just like they earlier determined that multi-cloud would far outpace standardization on any single cloud provider. Snowflake (28%), Databricks (29%), and Google Cloud’s BigQuery (27.6%) have nearly equal shares of adoption as enterprises move towards a heterogenous technology stack with Airflow as the connective tissue.
- Airflow is business-critical – Airflow isn’t just extremely popular with data teams (90%+ recommendation rate), it’s a core component of mission-critical data operations at leading enterprises. The majority of users in organizations with 50,000+ employees report driving business-critical workloads on Airflow.
- Airflow has extended beyond analytics – While Airflow remains a highly-effective framework for ETL and ELT–and it is still an excellent solution for advanced analytics and internal dashboards–it is rapidly evolving into a critical platform for deploying machine learning and generative AI solutions. Over 85% of surveyed users expect an increase in external or revenue-generating solutions built on Airflow in the next year.
- Orchestration-first organizations deliver production AI and ML – At a time when organizations in every industry are challenged with delivering a return on their AI investments, Airflow maturity and data orchestration foundations are key to getting AI and MLOps solutions in production. Among Airflow users with at least 5 years of experience, 30.6% run MLOps workloads on Airflow and 13.3% use Airflow for GenAI. Among Astronomer’s Astro customers, 55% use Airflow for ML and AI. Among customers that have used Astro for 2+ years, the number jumps to 69%.
The complete State of Airflow Report 2025 has more findings around the growing role of data teams, how large enterprises are leveraging heterogenous tech stacks, and key trends around data orchestration.
Also Read: The Role of AI and Machine Learning in Streaming Technology
[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]
Comments are closed.