# Ray normal DAG vs Compiled DAG

**URL:** <https://discuss.ray.io/t/ray-normal-dag-vs-compiled-dag/23247>\
**Category:** Uncategorized\
**Created:** [October 15, 2025, 1:36am UTC](https://discuss.ray.io/t/ray-normal-dag-vs-compiled-dag/23247 "2025-10-15T01:36:56Z")\
**Posts on this page:** 1\
**Showing post:** 1

<div class="post-metadata">

**Author:** ![Monil](https://avatars.discourse-cdn.com/v4/letter/m/ecb155/32.png) [@Monil](https://discuss.ray.io/u/Monil)\
**Post date:** [October 15, 2025, 1:36am UTC](https://discuss.ray.io/t/ray-normal-dag-vs-compiled-dag/23247/1 "2025-10-15T01:36:56Z")

</div>

Hello Community,

I am from offline inference platform team where we trigger standard Ray Batch Inference solution as suggested. Typically what we do is

1. we create a ray dataset with out custom datasource
2. We invoke map\_batches that takes an actor with bunch of init params.
3. We write data to our custom datasource.

I want to understand how normal Ray DAG would look like in the above flow vs how it would look like with Ray Compiled Graph? what is the main difference in the complied DAG?

---

_[View the full topic](https://discuss.ray.io/t/ray-normal-dag-vs-compiled-dag/23247)._