Future, Queueable, Batch — Key Differences
1. Future Methods
- Asynchronous, lightweight.
- Run in the background but cannot be chained.
- Limits: No monitoring, no state passing, can’t handle large datasets.
- Best for: Small tasks like sending email, making callouts, or updating a few records.
2. Queueable Apex
- Similar to future, but more powerful.
- Supports chaining (queue another job).
- Can pass complex data types.
- Still limited to processing data within governor limits (can’t handle millions directly).
- Best for: Medium-volume jobs or when you need chaining + flexible data structures.
3. Batch Apex
- Designed for large data volumes.
- Breaks records into manageable chunks (default 200).
- Can process up to 50 million records.
- Supports stateful operations (if you need to maintain state between batches).
- Best for: Heavy-duty data processing (millions of records).
Scenario: Processing 5 Million Records
- Future methods- Not suitable (can’t handle bulk).
- Queueable – Better than future, but still not scalable to millions. You’d hit limits quickly.
- Batch Apex – Best choice because it’s designed to process up to 50M records in chunks safely within governor limits.
To process 5 million records, I would use Batch Apex. Future and Queueable are good for smaller jobs, but they can’t handle millions of records because of governor limits. Batch Apex is designed for this use case—it processes data in chunks of 200 records by default and can scale up to 50 million records. This way, each batch stays within limits while allowing me to process the entire dataset reliably.