Feature
Snowflake Openflow
Move data from many sources into Snowflake with a managed integration service powered by Apache NiFi, ready for analytics and AI.
Build with powerful extensibility
Flexibly connect your most critical data to meet business demands.
Turbocharge ELT for AI
Unlock ELT pipelines to empower AI agents to make decisions at machine speed, leaving no data behind.
Scale with ease
Create scalable, enterprise-ready integrations with flexible deployment, automated observability and unified governance.
Change data capture
Capture database changes and put them to work for analytics and AI
Change data is hard to track across distributed systems in near real time. Snowflake Openflow captures inserts, updates and deletes.
Connect PostgreSQL, SQL Server and many more with prebuilt connectors that snapshot each table and keep it current.
Govern replicated tables with your existing Snowflake access controls.
Stream Oracle changes from redo logs via Oracle XStream with minimal OLTP impact, or bring your GoldenGate license.
DEPLOY CONNECTORS FASTER
Connect data sources faster and smarter with Snowflake CoCo and a guided wizard
Connect data from key sources to Snowflake without having to manage every setting or configuration yourself. Get your data flowing to Snowflake in less time and with fewer errors, using AI.
Control your pipeline so routine tasks like starting, stopping or rotating credentials never slow you down.
Monitor exactly which connectors are deployed and what each one is doing at any point in time.
Discover your data estate automatically so you can know what’s connected and where your data lives.
CoCo takes it further. Install, troubleshoot and manage connectors in plain language.
A step-by-step wizard walks you through connector configuration so your first pipeline can run in minutes.
Enhance extensibility
Connect critical data to power business demands
Facilitate open and interoperable architecture. Move data to data lakes and lakehouses, and easily adapt to new industry standards like Apache Iceberg.
Unify structured and unstructured, batch and streaming data movement into one platform with connectors across SaaS, OLTP databases, event streams, file storages and much more.
Simplify key data integrations with dozens of out-of-the-box connectors and hundreds of processors.
Build custom connectors that run on Snowflake's managed platform with an inherently open, extensible foundation.
Empower AI
Fast-track AI deployment with ELT
Work with multimodal data directly from the source, including SharePoint, Slack, Box and Google Drive.
Comply with source system access control list (ACL) and carry consistent user permissions with the source.
Communicate in near real time and bidirectionally with streaming and continuous data movement (via HTTP listeners and publishers).
Built-in Snowflake Cortex AI capabilities can parse and pre-process the unstructured files before writing to Snowflake — all in one platform.
Simplify operations
Scale with confidence
You decide where to deploy integration pipelines: via Snowflake-managed infrastructure or bring your own cloud (BYOC) within your VPCs closer to where your source systems are.
Reach on-premises databases and private cloud sources with Data Connectivity Proxy (DCP), a lightweight agent that connects outbound to Snowflake with no inbound firewall rules required.
Know where your data and pipelines come from — and what happens to them — every step of the way thanks to full data lineage.
Get enterprise-grade governance, including advanced role-based access control (RBAC).
Snowflake Openflow Overview
Powered by open source
Openflow supercharges Apache NiFi with enterprise-ready governance, security, deployment flexibility and observability.
Openflow connectors
Highly curated, managed, secured and governed against source systems, Openflow connectors are extensible and customizable with processor components.
Openflow Service and API
Openflow service and API run in Snowflake as a managed service, allowing you to provision and manage data pipelines easily with an intuitive UI. Get a unified user experience with deep observability via the Connector Dashboard, including real-time monitoring and alerts, DAG visualization and refresh history information.
Deployment and runtime
Run Openflow connectors and pipelines in the environment of your choice. Execute in your own VPC via Bring Your Own Cloud (BYOC) deployment on AWS, or choose Snowflake-managed deployment, available on AWS, Azure and GCP.
Mercury Turns Fresh Data into Faster Product Decisions with Snowflake Openflow
With Snowflake Openflow, Mercury can trust the numbers behind its decisions, ship product updates faster and put engineering time back into building the business.
- 1-3 Minutes of data latency to Snowflake, down from 30-60+ minutes
- 0 Replication incidents since the migration
Openflow Partners
Snowflake Openflow
Frequently Asked Questions
Find answers here about Snowflake Openflow features, Apache NiFi, and how it can streamline your data pipelines.
Snowflake Openflow is a managed integration service built on Apache NiFi. It is designed to connect critical data sources, supporting structured and unstructured data. Run Openflow in your own cloud environment for control, or choose Snowflake-managed deployment to reduce operational burdens.
Snowflake Openflow is built on Apache NiFi, leveraging its robust data flow capabilities. This means you can use familiar NiFi processors and controller services alongside Snowflake-specific components. These components include advanced governance and control, observability and the option to run and deploy integrations in your cloud or via Snowflake Deployment with Snowpark Container Services. Both of these services are managed by Snowflake.
Openflow is designed to handle a wide array of data, including structured, semi-structured and unstructured (text, images, audio, video, sensor data). It offers connectors for various SaaS platforms (e.g., Google Ads, Meta Ads, SharePoint, Box), databases (Oracle, MySQL, PostgreSQL, SQL Server), streaming services (Kafka, Kinesis), and more.
Openflow facilitates the near real-time ingestion of multimodal and unstructured data from sources like Google Drive or SharePoint. The connectors provide built-in capabilities to extract, preprocess (using Snowflake Cortex LLM functions), load and activate unstructured data directly in the ETL pipeline. This makes the data readily available for AI processing and applications, such as building "chat with your data" experiences using Snowflake Cortex AI.
Snowflake Openflow is a managed service that offers customers options to run their deployment via Bring Your Own Cloud (BYOC) or via Snowflake deployment. BYOC runs the integration within your own cloud environment — currently available in AWS commercial regions and deployed in your Virtual Private Cloud (VPC). This gives you more control over your data and network. Snowflake manages the Openflow service itself, including the UI, observability and APIs for deploying and managing runtimes. With Snowflake deployment (generally available on AWS, Azure and GCP), all deployments and runtimes are managed by Snowflake.
Snowflake Openflow BYOC is generally available in all AWS commercial regions, and Snowflake Openflow deployment is generally available on all AWS, Azure and GCP commercial deployments.
Yes. The Openflow Connector for Oracle reads Oracle redo logs through XStream and streams changes into Snowflake in near real time, with no software installed on the Oracle servers. You can license XStream through Snowflake or bring an existing Oracle license.
