Synchronous vs. asynchronous integration patterns

Area

Synchronous

Asynchronous

Mechanism

Data swaps hands synchronously, sparking actions via services.

Data moves on its own, either on a set schedule or bit by bit using messaging patterns.

Explanation of Mechanism

In a synchronous mechanism, the data exchange is immediate, and the sending system waits for a response from the receiving system before proceeding. This is ideal for real-time operations where the response is needed instantly.

Asynchronous mechanisms allow data to be transferred without waiting for immediate processing. The data can be sent in batches or in chunks over time, allowing the systems to process it at their own pace.

Trigger

User action or system event.

Scheduled or user-initiated. Can wait for downtime or quiet periods.

Explanation of Trigger

In synchronous systems, the trigger could be an immediate action, such as a user clicking a button or a system automatically updating a record, prompting an instant response or data exchange.

Asynchronous triggers can be based on a schedule (e.g., nightly data backups) or initiated by users but executed when the system is less busy, such as during non-peak hours, to avoid system overload.

Pros

Quick back-and-forth. Up-to-the-minute values and information.

Systems are not too reliant on each other, making for a sturdy solution. Balances loads over time. Can be almost real-time.

Explanation of Pros

Synchronous systems are great for tasks that require immediate feedback, ensuring that all systems have the most current data at all times.

Asynchronous systems reduce dependency between systems, enhancing reliability. They also allow better load balancing, as data can be processed during less busy times.

Cons

Only small payloads. Can lead to systems being too dependent on each other. Sensitive to delays.

There is a wait for responses and updates across systems.

Explanation of Cons

Synchronous systems may struggle with large data transfers, and their performance can degrade if there are delays. They also create dependencies, meaning if one system fails, it can impact the others.

In asynchronous systems, data transfer and processing are not instant, meaning there may be delays before all systems are updated. This might not be ideal for time-sensitive tasks.

Blocking Pattern

A synchronous pattern is a blocking request and response pattern, where the caller is blocked until the recipient has finished running the Excel integration and gives a response. 

An asynchronous pattern is a non-blocking pattern, where the caller submits the request and then continues without waiting for a response.

Explanation of Blocking Pattern

In a synchronous, blocking pattern, the system waits (or is 'blocked') until the response is received before moving on to the next task. This is essential when the next task depends on the immediate result of the current one.

In asynchronous, non-blocking patterns, the system can continue processing other tasks while waiting for a response. This increases efficiency but may not be suitable when immediate results are needed for subsequent actions.

Scenarios

Consider using OData for synchronous integration scenarios.

Consider Batch data API for asynchronous integration scenarios.

Explanation of Scenarios

OData is well-suited for synchronous operations where real-time data retrieval and updates are crucial, such as retrieving live data for display in user interfaces.

Batch data APIs are ideal for asynchronous operations, where data can be processed in bulk during off-peak times, such as in overnight data synchronization or reporting processes.

Use when

You cannot do without real-time information.

Most recommended integration patterns and technologies are asynchronous but can feel like real-time.

Explanation of Use when

Synchronous integration should be used when immediate feedback or real-time data is critical, such as in live monitoring or transaction processing systems.

Asynchronous integration is typically preferred when systems can tolerate slight delays, allowing for greater flexibility and efficiency, such as in reporting or batch data processing.