Next Destination for Geo based on Customer Feedback

Background

There have been a few recent (and less recent) issues about the long term goal for Geo.

• #8353 (closed) - Mid/long term direction for Geo using federation (This issue contains a number of other issues talking about technical goals for Geo).
• #8411 (closed) - Secondary first MR's

It is important for us to consider what our long term goal is so that we make good short term decisions, and to do this we need to spend time researching our options.

Customer Centric Research

Each of the previous issues focuses on the technical ability of the feature, and what that affords the customer. For this issue I would like us to think about this the other way around. What do our customers want, and how will we build that.

Selective Sync of Projects : some customers with many projects want the ability to be able to sync only certain projects to certain nodes. They have legal requirements that data for certain projects is not physically located outside of the USA. Yet they still need to have a Geo node closer to their users.

Writable Nodes : customers want to be able to write to the nodes to reduce latency even further.

Transparent Setup : customers want to be able to congifure their client the same way globally and let their operations teams orchestrate the nodes transparently to the users.

These are the first few that come to mind, I will add others as I find them. (And I'll find the issue links).

Outcome for this issue

Research like this has the potential to be long running. This issue is to set up the research project and create issues for the various avenues of enquiry.

The outcome for this issue is to create a set of issues that describe the research required for each topic. This issue is about breadth, and the set of issues that needs to be created is about depth. If possible, each issue should take no longer than one milestone to research.