IPaaS vs. API management: Why the enterprise needs both Azure Logic Apps: How it compares to AWS Step Functions

AWS launches Amazon AppFlow for SaaS integration

The newly launched Amazon AppFlow gives AWS an answer to the likes of Microsoft's Azure Logic Apps, for tying together its native services with SaaS application data.

AWS has launched Amazon AppFlow, a managed integration pipeline that moves data securely between SaaS applications and native services, including S3 storage and the Redshift data warehouse.

The service gives customers an automated way to bring data from SaaS apps into AWS, where it can be analyzed and used to create visualizations and even new applications. AppFlow's goals are similar to Microsoft's Azure Logic Apps, although the latter have an emphasis on workflow automation as well as data integration.

AWS already had services with similar intent to AppFlow, such as EventBridge, which focuses on real-time data ingestion scenarios. EventBridge is aimed at developers building new applications with AWS services such as the Lambda serverless platform.

AppFlow is a welcome addition to the mix, said Ryan Marsh, a DevOps and serverless expert at TheStack.io in Houston.

"This will continue to make it easy to build low-code/no-code serverless apps on AWS," Marsh said. "If you combine it with AWS Step Functions, which I didn't see in the docs but would expect to be a feature at some point, you could build some impressive integrations." Step Functions, like its predecessor Simple Workflow Service, provides workflow automation tooling for developers.

This will continue to make it easy to build low-code/no-code serverless apps on AWS.
Ryan MarshDevOps and serverless expert, TheStack.io

At launch, Amazon AppFlow supports more than a dozen data sources. They include Salesforce, Datadog, Amplitude, ServiceNow, Dynatrace, Google Analytics, Infor Nexus, Marketo, Singular, Snowflake, Slack, Trend Micro, Veeva and Zendesk. While AWS will undoubtedly add many more over time, that starting lineup represents a broad cross section of application domains.

AppFlow's prebuilt, supported integrations save customers time and help them avoid risk, in contrast to writing integrations on their own or conducting manual import-exports, AWS said in a blog post. AppFlow also doesn't require writing any code, making it amenable to power users as well as developers.

The integrations are bidirectional, meaning that applications can also ingest data from AWS services. Data is encrypted in motion by default, but an additional layer of security is applied if the SaaS provider has integrated with AWS' PrivateLink interconnect, which doesn't use the public internet. Data transfers can be scheduled, triggered by events or generated on demand, according to the post.

Major data integration methods
Major data integration methods

AppFlow is accessed through the AWS Management Console. From there, users select data sources, map out the information they want, and if desired or required, apply data transformations such as merges and filters.

It is available now in North America, Asia-Pacific, Europe and South American AWS regions. It's priced based on "flow runs," which AWS defines as "a call to the source application to transfer data to a destination." Each call costs $0.001 and individual AWS accounts are limited to 10 million flow runs per month. Data processing charges also apply if customers choose to make transformations, with costs starting at $0.02 per GB. AWS also charges standard fees for reads and writes from its native services.

Dig Deeper on Cloud app development and management

Data Center