| status | date | deciders |
|---|---|---|
accepted |
2023-09-15 |
Idealos |
The technical issue at hand concerns the implementation of a Reservation Updates Scheduler to manage and update bookings that do not support event subscription. We need to determine the approach for scheduling and processing updates for these reservations to ensure data accuracy and timeliness.
Our system processes a wide range of reservations, some of which are made through booking channels that do not offer event subscription capabilities. As a result, these reservations cannot be updated in real-time through event-driven mechanisms. To address this limitation, we need to implement a Reservation Updates Scheduler to periodically fetch and update reservation information.
After thorough analysis and consideration, it has been decided to implement a Reservation Updates Scheduler to handle bookings without event subscription support. This scheduler will be responsible for the following tasks:
-
Regular Polling: The scheduler will periodically poll the external booking sources that do not support event subscription to check for updates to reservations. The polling frequency will be determined based on the criticality of updates and the policies of the booking sources.
-
Data Synchronization: Upon detecting updates during polling, the scheduler will retrieve the latest reservation data and synchronize it with our system's database. This process will include updating reservation details such as dates, customer information, and booking status.
-
Error Handling: The scheduler will be equipped with robust error-handling mechanisms to address any issues that may arise during the polling and synchronization processes. This includes handling network failures, temporary unavailability of external sources, or data format inconsistencies.
Implementing a Reservation Updates Scheduler for bookings without event subscription support has several implications:
-
Latency: Updates for reservations processed through this method will not be in real-time, leading to potential latency in reflecting the latest information in our system.
-
Resource Consumption: Regular polling may consume network bandwidth and processing resources, particularly if a large number of reservations need to be synchronized.
-
Data Consistency: Care must be taken to ensure data consistency and avoid conflicts when multiple updates occur for the same reservation during polling.
-
Monitoring and Alerting: Robust monitoring and alerting mechanisms should be in place to track the status of the scheduler, detect failures, and trigger corrective actions promptly.