Nvidia provides a special library the DALI library (Data Loading Library) to help loading data on GPUs, called NVIDIA Data Loading Library (DALI) that . DALI can speedup deep learning using both GPU and CPU a mix of GPUs and CPUs for data loading process. DALI is capable of overlapping the training and preprocessing stages; thus, reducing the latency and training time.    Another advantage is that it provides an a unified data loading interface to Deep learning frameworks. Different frameworks have different data loading classes, and it is not possible to use a data loading class from one framework with another one. Dali DALI provides a solution in form of a data loading class that can be used with all well-know known Deep learning frameworks. 

Next sectionBelow, we present a Jupyter notebook that loads and process data with DALI. Follow the steps below to launch the notebook in the NCI ARE interactive environment,  named ARE. The notebook is designed for people who may not be familiar with DALI, but wants to learn more with the help of practical examples.

Launch ARE notebook

1. goto go to the url: https://are.nci.org.au/ and login. 
2. Open the web form to launch a new Jupyter Notebook.
3. Choose "gpuvolta" queue and compute size "4gpu". Also add the necessary storage option: gdata/dk92+gdata/wb00  (You can also append extra storage space as shown below) 
   
   
   
   
4. Goto Go to the "Advanced section..." and choose the following:
   Modules directories: /g/data/dk92/apps/Modules/modulefiles
   Modules: NCI-ai-ml/22.11
   
   
   
   
5. Wait for the JupyterLab to start.

...