Survey and EHR data in All of Us are often manageable in size and the data costs associated with them are usually low (especially if you are following All of Us' optimization tips). However, genomic data and wearable data can have significant data costs due to the size of the datasets.
If you are using wearable data for your research, consider using daily averages instead of minute-level data, as the continuously collected data files can be extremely large and have high data costs associated with them. Similarly, make sure to test your workflows for genomic data on small subsets before moving on to your full analysis.
No. Computational, data, and storage costs will be billed against the workspace owner's account.
However, there is a possible workaround if you have multiple collaborators on a project who each have free credits. When the initial workspace owner starts to run low on credits, you can duplicate your current workspace and assign the new copy to a different owner on the team. If you do this, be sure to re-add your collaborators to the copied workspace and make sure to transfer your storage buckets.
While the All of Us data is free to use, there are charges incurred for the use of the cloud computing environments in the All of Us Researcher Workbench. Three factors influence how much you will be charged:
You can estimate your data costs using this basic formula:
Time: the amount of time measured in minute increments
Read more about cloud computing and computation costs on the All of Us user support website:
Currently, new users receive $300 in free credits for the Researcher Workbench. These credits expire 120 days after completing the AoU Responsible Conduct of Research Training and signing the Data Use Code of Conduct. Once you have used your free credits, see the links below for guidance on setting up a billing account. If your research is NIH-funded, consider applying for funding through the STRIDES Initiative (linked below) to cover additional computational costs.
All of Us has provided a great deal of guidance on using the Researcher Workbench efficiently in order to optimize compute time and reduce costs. Check out the articles below for more information: