Introduction: We are announcing the availability of the grip database, effective [release X]. This decision is based on its ability to handle cohort queries as well as grip (and bmeg) capacity to handle meaningful scientific queries.
Reasoning: After careful review and consideration, it has become evident that the psqlgraph (sheepdog) database has not gained significant traction or usage since its introduction. It is not used by the front-end-framework and is limited in its ability to support deep analysis queries
Modification to aced_submission meta graph load:
Update functionality to no longer load the PostgreSQL database. Instead, modify it to submit metadata to the Grip loader. Ensure all dependencies and configurations are adjusted accordingly.
Enhancement in Grip Query Engine:
Implement a mechanism in the Grip query engine to:
Call authz/mapping to fetch current user's authorizations.
Cache the authorization results for the (lifetime of) the token to optimize performance and reduce latency.
Query Engine Filtering Enhancement:
Automatically integrate a "WHERE" clause in the Grip query engine to enforce vertex filtering based on auth_resource_path attributes. Ensure that vertices must match entries in the authz/mapping for access authorization.
Deployment via Helm Chart:
Develop a Helm chart specifically for deploying MongoDB and Grip components. Ensure the chart includes configurations for scalability, resilience, and ease of deployment in different environments (rancherdesktop -> production).
Integration of Grip Client in Jupyter Notebooks:
Install and configure the Grip client within Jupyter Notebooks. Provide clear documentation and guides for users to seamlessly integrate and utilize Grip functionalities within their notebook environments.
Introduction: We are announcing the availability of the grip database, effective [release X]. This decision is based on its ability to handle cohort queries as well as grip (and bmeg) capacity to handle meaningful scientific queries.
Reasoning: After careful review and consideration, it has become evident that the psqlgraph (sheepdog) database has not gained significant traction or usage since its introduction. It is not used by the front-end-framework and is limited in its ability to support deep analysis queries
Modification to
aced_submission meta graph load
:Enhancement in Grip Query Engine:
authz/mapping
to fetch current user's authorizations.Query Engine Filtering Enhancement:
auth_resource_path
attributes. Ensure that vertices must match entries in theauthz/mapping
for access authorization.Deployment via Helm Chart:
Integration of Grip Client in Jupyter Notebooks: