Docker containers offer developers with the agility and flexibility of replicating the production/test setup in their development environment. So now, developers can develop features, unit test, fix issues in their local setup before pushing it to the test/production environment. Since containers are state-less, it can be moved anywhere easily, say within datacenter or into clouds etc. But in most cases, containers has to access/modify data that traditionally lives in a centralized storage. For example, one of the popular container stack LEMP, needs to access MySQL database. In order to access these data, Docker developed the concept of volume plugins, which can be associated with a container when it is created. Now the container can move to a different location as long as the volume plugin works there as well. This solves the problem in production/test cases, but if developers have to access the centralized data, that again restricts their flexibily. At the same time, with ever increasing size of data, it is not possible to give each developer a separate copy of the data (say database). That would be prohibitively expensive. To solve this, we give you "Horcrux"...