AppFlow exchanges data between supported SaaS applications and AWS services. Choose the connector direction and trigger, then define which records and fields to transfer.
SaaS data integration
Amazon AppFlow transfers data between supported applications and AWS services using configured connectionsStored application access settings used by a flow to reach its source or destination. and mappings. Software as a serviceAn application hosted and operated by its provider that customers access as a service. (SaaS) means an application hosted and operated by its provider. A flowA configured data transfer between a source and destination. is the transfer definition; its connectorAn integration that defines how AppFlow communicates with a supported source or destination. determines how AppFlow communicates with the source or destination.
How it works
- Choose a supported source and destination, such as Salesforce and S3. Check direction support for the connector. A connection holds application access settings; its connector profileA reusable connection resource that a flow creator must have permission to use. must be usable by the identity creating the flow through
UseConnectorProfile. - Select a triggerThe condition that starts a flow run., the condition that starts a run, using the table below. Scheduled flows can copy all records or just records added or changed since the last successful run.
- Map source fields to destination fields in the console or with a CSVA text format that separates field values with commas. file, a text file containing comma-separated pairs. Add value changes such as maskingChanging field values to conceal their original contents during transfer., validation rules for bad data, and filters selecting the records to transfer.
- For incremental schedules, select a timestamp field that captures the required changes. Use an offset if the source records changes late, so the transfer window allows for that delay.
- AppFlow encrypts connection credentials with KMS and stores them in Secrets Manager. Data transfer uses TLSA protocol that encrypts data while it travels between endpoints. 1.2, which encrypts data moving across the network. PrivateLink can keep the actual transfer off the public internet for supported SaaS connections.
| Trigger | Starts when | Constraint |
|---|---|---|
| On demand | A user manually runs the flow | Run it for each transfer |
| Event | The source produces the chosen change event | Source must support change events |
| Schedule | The recurring schedule runs | Frequency depends on the source; choose full or incremental |
Exam cues
- If the question says move Salesforce records to S3 without custom integration code, think AppFlow, because it provides managed application connections and data flows.
- If the question says copy new and edited records periodically, think incremental scheduling with an appropriate timestamp field, because Created Date alone misses edits.
- If the question says mask fields and select records during transfer, think AppFlow value modifications and filters, because those operations are part of flow configuration.
- If the question says replicate ongoing database transaction-log changes, think DMS change data captureCapturing ongoing source database changes for replication., because that mechanism reads database logs rather than using a SaaS flow connector.
Remember
- The first scheduled run pulls 30 days of past records. For another initial range, use a filtered on-demand run before switching to incremental scheduling.
- Incremental offset defaults to zero and must be less than ten hours.
- Multiple record filters combine with AND: records must satisfy all those filters.
- Private transferFlow data movement through PrivateLink endpoints without exposing that data path to the public internet. requires a supported connector and a PrivateLink-enabled SaaS account.
Trap: A private AppFlow data transfer does not make every console API call private. Calls that list source objects and fields still use public endpoints.
Trap: A listed connector is not a promise of both source and destination support or every trigger type. Check its supported direction and events.
The selected trigger starts a flow using the configured application connection.
AppFlow applies configured mappings and filters and transfers the selected records to S3.
From the AWS docs (19)
- AppFlow is a managed integration service for exchanging data between supported SaaS applications such as Salesforce and AWS services such as S3 and Redshift without custom flow code.
- AppFlow sources and destinations depend on the connector; an application is not necessarily supported in both directions.
- Creating a flow selects source and destination connections and requires permission to use the selected connector profile through UseConnectorProfile.
- AppFlow supports on-demand, event-triggered, and schedule-triggered flows.
- Event-triggered flows require a source SaaS application that supplies change events.
- A scheduled flow frequency depends on the source application capabilities.
- A scheduled full transfer copies a snapshot of all source records at the time of the run.
- Scheduled incremental transfer copies new or changed records since the last successful run; the selected source timestamp field determines which changes are identified.
- Choosing a Created Date source timestamp for incremental transfer selects newly created records but does not select edits to existing records.
- The first scheduled run pulls 30 days of past records; an on-demand filtered initial run can load another range before switching to an incremental schedule.
- An incremental schedule can use a time offset to accommodate delayed source timestamps; the default offset is zero and its value must be less than ten hours.
- AppFlow supports field mapping through the console or an uploaded CSV file pairing source and destination fields.
- Flow configuration can concatenate fields, mask or truncate values, and validate fields with a configured action for bad data.
- AppFlow record filters select which records to transfer; multiple filters combine with AND logic.
- AppFlow stores application connection credentials in Secrets Manager encrypted using KMS keys.
- AppFlow secures transferred data in transit with TLS 1.2 and supports AWS managed or customer managed KMS keys for flow data encryption.
- Private flows use PrivateLink for the data transfer with supported SaaS applications; the SaaS account must be enabled for PrivateLink.
- Even for a private flow, console calls that list source objects and fields use public endpoints; the flow data transfer uses PrivateLink endpoints.
- AWS DMS supports ongoing database change data capture by reading database change logs, which is a different use case from SaaS connector flows.
Flashcards
Nothing to review here right now. Pick other topics or another filter.
·
All 9 cards as a list
A team needs to ingest Salesforce data into Amazon S3 on a recurring schedule without custom integration code. What can it use?
Amazon AppFlow can transfer data from supported SaaS applications to AWS services on a schedule.
A team needs a recurring Salesforce to S3 ingestion of new and changed records. Which AppFlow transfer mode fits?
A scheduled incremental transfer copies new or changed records since the last successful run.
A team wants Amazon AppFlow to ingest data when a SaaS record changes. What must the source support?
The source SaaS application must supply change events for an event-triggered flow.
A team needs a complete snapshot of SaaS records in Amazon S3 every day. Which AppFlow ingestion pattern fits?
Use a scheduled full transfer. Each run copies a snapshot of all source records.
A team needs to start a SaaS data transfer manually when requested. Which Amazon AppFlow trigger fits?
Use an on-demand flow.
Which Amazon AppFlow trigger options support different data ingestion frequencies?
Flows can run on demand, on a schedule, or when a supported source supplies change events.
A team plans frequent scheduled SaaS ingestion with Amazon AppFlow. What determines the available schedule frequency?
The source application capabilities determine the available frequency.
An Amazon AppFlow scheduled ingestion must include edits to existing SaaS records. Is Created Date alone a suitable incremental timestamp?
No. Created Date selects newly created records and misses edits to existing records. Choose a timestamp that captures the required changes.
An Amazon AppFlow flow must regularly transfer only new or changed SaaS records. Which transfer option fits?
Use a scheduled incremental transfer. It copies records added or changed since the last successful run, based on the selected source timestamp field.