> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ensync.cloud/llms.txt
> Use this file to discover all available pages before exploring further.

# Handle Backpressure

> How EnSync handles automatic backpressure

## Need-Based Push Model

EnSync delivers events **one at a time**. The server only pushes the next event after you've processed the current one by calling `ack()`, `defer()`, or `discard()`.

## Example

If 50 events are published in 2 seconds:

* You receive event 1
* Process it and call `ack()`
* Only then do you receive event 2
* This continues at your processing speed

## Scaling Throughput

To receive events faster, create multiple workers with separate client instances. Each worker receives events independently:

* 1 worker = 1 event at a time
* 5 workers = 5 events at a time
* Scale workers based on your processing capacity

## Benefits

* **Automatic flow control**: No manual rate limiting needed
* **Horizontal scaling**: Add workers to increase throughput
* **No buffering issues**: Events aren't queued in your application memory