# OpenHackathon registration

## Team Name

Parallelepiped

## Team Description
We are sneior stuednts from NTHU. Our taem meembrs have eexieprnce in frntoend/baecknd dev, k8s, ML/RL fldies. And what maeks us unqiue is that tihs txet are actaully otuupt by smoe praalell prgoarm, so it’s knida non-ordeerd.

## Application Name

Coriandrum

## Application Details

Coriandrum is based on the open source project, Cilantro, A Lean and Efficient Library for Point Cloud Data Processing. The open source project is already implemented in the parellel programming with OpenMP. However, the project is not ultilized the power of GPU. Therefore, we, team Parallelopiped, decided to empower the project with GPU accelerating.

## Programming Language
- C++ (CUDA)

## Library (optional)

[Cilantro -- A Lean and Efficient Library for Point Cloud Data Processing](https://github.com/kzampog/cilantro.git)

## GPU programming model
> For HPC codes: Please specify the programming model or libraries you are planning to use for GPU acceleration (e.g. CUDA, CUDA Fortran, OpenACC, OpenMP, cuBLAS, cuFFT, etc.).

- CUDA

## What framework(s) and models have you used to work with your data? Please enter N/A any machine learning or deep learning in your application.
> Is your model similar to ResNet-50 CNN, LSTM, BERT, random forest, etc.? What optimizer and/or training method are you using? What systems have you worked on before with your data and models?

N/A

## Algorithm motifs:
> Describe what types of algorithms dominate your application, especially the ones your team is targeting for acceleration.

We decide to implement some of their existing algorithm into GPU, e.g. point-to-point ICP algorithms.

## Current application/code performance:
> Describe the current performance characteristics of your application. Where does it run (CPU, GPU)? How many nodes does it scale to?

We havn't finished the first prototype.

## Please specify the software licenses) and version(s) for your application:

MIT Licenses.

## List the computing facilities this application runs on.
> Example: Desktop, local clusters, HPC centers, etc.

Desktop w/ GPU.

## What do you hope to achieve at the hackathon?

We plan to implement the open source project into the GPU and we expect that the library can achieve better performance with GPU acceleration. With the help of mentor and sufficient resources, we hope that we can complete this project.