# [DaCe] GPU target for GT4Py backend

###### tags: `cycle 18` <!-- Change to the current cycle number -->

- Shaped by: Edoardo
- Appetite (FTEs, weeks): 2
- Developers: <!-- Filled in at the betting table unless someone is specifically required here -->

## Problem
<!-- The raw idea, a use case, or something we’ve seen that motivates us to work on this -->
In cycle 17 an issue was found in DaCe for lowering of SDFG to CUDA code: https://github.com/spcl/dace/issues/1388

As the description of the GitHub issue says, there is a workaround but not a proper fix. This issue is the first step in fully achieving GPU support in the GT4Py DaCe backend.

Next step, which is also in the scope of this pitch, is to ensure that DaCe auto-optimize heuristics work for the GPU target device.

Finally, the DaCe backend should be properly integrated with the GT4Py frontend for storage of fields in GPU memory and runtime selection of CPU/GPU target.

## Appetite
<!-- Explain how much time we want to spend and how that constrains the solution -->
The appetite of 2 weeks does not include the learning activity for DaCe ramp-up.

## Solution
<!-- The core elements we came up with, presented in a form that’s easy for people to immediately understand -->

## Rabbit holes
<!-- Details about the solution worth calling out to avoid problems -->

## No-gos
<!-- Anything specifically excluded from the concept: functionality or use cases we intentionally aren’t covering to fit the ## appetite or make the problem tractable -->