In this chapter, we expand the discussion on sharding by planning for deployments with very large datasets or extensive write workloads. We review challenges for sharding in these cases and explore methods to avoid or mitigate problems as the data grows.
Since we are working with shards, all examples and commands shown in this chapter must always be run through a mongos instance and never directly on the shards themselves.
Indications for sharding
There are a few different options for scaling depending on the size of the data and the access patterns. If you have a “small” database ...
