Problem:

Currently, the only way the customer can have a look at the data and results of any analysis is via customer APIs/UIs somebody needs to build first. An alternative might be to allow users to have a subset of the data-lake with their data available for their analytics and reporting purposes. The data would contain the warehoused version of their data as well as results from any analysis put on top of it.

Target persona:

• ISV or a large customer wanting to further process CCX data
• ISV or 3rd party wanting to integrate additional views on CCX data into their monitoring tools 

Goal:

Phase 0

In the first phase, there would be no guarantees over the data schema (so that we don’t have to solve the schema versioning from the very beginning in case some changes would be needed during initial phases): the main focus would be on full isolation of the data so that the customer would get exclusively only to the data he owns.

Phase 1: 

A data-lake for an ISV partner could be built in a similar way: there will be a definition of subset of the full data-lake to be generated for each partner account, including only data the partner should have access to.

The same mechanism could be also used to better control the access to the data internally: instead of giving everyone access to the full data set, the system could provide just the subset of data necessary for the job, keeping the data more secure as well as the user having more efficient access (not having to use the data they would not need).

Unknowns:

Given not every customer would be probably using this functionality, it should be provided on demand and being disabled by default.