-2

Please could You share a study- or research-paper on how to implement “trunk-based”/master-branch development (as opposed to have long-lived sub branches) in a micro-service based environment ?

Preferably a study giving business case ($$$) justifications and proofs of benefits of such approach.

Which can convince developers and managers alike that it’s important to strive for short-lived branches for example using “feature toggles”.

As opposed to having long-lived (sub)branches that is difficult to merge back into master, and which makes automation testing hard due to not testing “the (main)trunk/master” sufficiently often.

Furthermore it would be interesting if the paper discuss, common pitfalls, pros and cons and (practical) obstacles in implementing "trunk-based" development for example how to handle changes in contract versions between micro-services(teams), for example using feature toggles between micro-services when trunk/master is updated continuously from many micro-teams?

If the paper mention PACT or similar framework would be a additional benefit to convince management in the right direction.

Thank you!

J_A_X
  • 12,857
  • 1
  • 25
  • 31
zombieboy
  • 116
  • 1
  • 10
  • According to our [on-topic](https://stackoverflow.com/help/on-topic) guidance, "**Some questions are still off-topic, even if they fit into one of the categories listed above:**...Questions asking us to *recommend or find a book, tool, software library, tutorial or other off-site resource* are off-topic..." – Robert Columbia Nov 09 '17 at 13:24

1 Answers1

0

With friends a put together a site: https://trunkbaseddevelopment.com that has a lot of the alternate choices, but you were after case studies. Here's one - https://devops.paulhammant.com/bofa-platform-case-study/. That was ten years ago, and a bunch of my work is explaining that to corporates, and which steps to take first on a multi-month series of steps to CD.

paul_h
  • 1,859
  • 3
  • 19
  • 27