Investigate how to handle HL7 integration into the API endpoints.

https://aws.amazon.com/blogs/compute/building-a-serverless-tokenization-solution-to-mask-sensitive-data/

Contact Philip about individual vs company contracts.

AWS recently launched

https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/s3_bucket

Using this for help - https://github.com/spe-uob/fhir-works-on-aws-deployment

1. Implement the Lambda,
2. Implement Cognito, S3 and API Gateway

in Serverless. If you want to rewrite it then fine but decide among your pair which language to use for the lambda: Python, (Type|Java)Script, Java, Golang. They've used TypeScript in that example so you can just copy it if you prefer. It really isn't critical so keep it simple.

3. Test it, write unit tests (unless you copied it because that has already tests) and usage instructions in readme so the data sim team can use the endpoint once we give them the credentials and url

Set up fossa as an open-source management service.

Fill out a template then create Kanban tasks for each section so that team can work on documentation this week.

Investigate thoroughly the implications for making this portable. To be honest, trying too hard to avoid vendor lock-in increases the technical debt substantially.

This is an issue that many organisations still struggle to navigate but they have the financial resources to hire DevOps and Infra teams. We are a small team expected to deliver a commercial production system in not very long and with a lack of experience that I think we risk over-engineering, over-promising and under delivering were we to to implement a full multi-cloud solution that is scalable. So what I want us to identify is where the pluggable components are, and how easy that transition is for a later product. This will all need to be documented thoroughly for the stakeholders.

I'm going to assign myself this but make a note of the issue for everyone's reference.

Investigate the features, pricing and dependencies. Preferred solutions are serverless. Read some documentation as well to see how they work in deployment.

e.g. QuickSight, Tableau

this comes from Redshift data marts that were curated using EMR

1. https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/glue_catalog_database
2. https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/glue_crawler

What we are building in the workshop on Friday is the following transformation from our DynamoDB to an S3 bucket in Apache Parquet format with Hive metastore data in Glue Data Catalog.

https://registry.terraform.io/modules/terraform-aws-modules/redshift/aws/latest
https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/redshift_cluster

https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/emr_cluster

Explain the proposed AWS integration for trusted research environment, federated querying and business insights.

• EMR Notebooks https://aws.amazon.com/emr/features/?nc=sn&loc=5
• Redshift https://aws.amazon.com/redshift/features/
• QuickSight https://aws.amazon.com/quicksight/