Transaction Isolation | Backendless Database Training Course (pt. 45)
In this video, we will dive into transaction isolation which is important to understand when building logic or writing code that uses transactions. Transaction isolation is relevant not only to Backendless but with many databases that support transactions.
In order to provide consistency and integrity of your data, transactions become quite useful. If an application has a lot of concurrency, a lot of users hitting the database at the same time and multiple users modifying the same database record concurrently, isolation ensures data integrity.
In this video, we will review all the models of isolation of transaction isolation that are present in Backendless. We will review the isolation models that are applicable whenever creating transactions and creating backups.
The isolation models are:
Read uncommitted: This isolation allows us to view data that is still uncommitted. We could see the effects of transactions even before they are executed.
Read committed: This isolation allows us to view only data that is committed data. It prevents dirty reads.
Repeatable read: This isolation mode is the default mode, and it prevents dirty reads and non-repeatable reads.
Serializable: This isolation mode is the most isolated and the most expensive in terms of performance. It provides complete isolation of all transactions.
Understanding the different isolation modes is essential when designing your transactions. Although repeatable read works for most cases, it is always good to consider your application's specific requirements for how data needs to be handled.
Have questions? Post them in the comments below or get in touch at:
Support forum: https://support.backendless.com/
Slack channel: https://backendless.slack.com/
Telegram channel: https://t.me/backendless_chat
Start with Backendless for free today to follow along with this course: https://develop.backendless.com/regis...
Happy Coding!
#database #api