Hashed sharding causes MongoDB to create a hashed index on the value of the shard key, rather than using the key as is. Hashed sharding tends to produce a more even distribution of data within the sharded cluster, and is an ideal solution if the shard key has high cardinality or is a field that automatically increments. Because the hashed value is used to perform distribution rather than the actual value of the shard key field itself, a shard key that automatically increments does not cause MongoDB to eventually dump all data into the last shard (as discussed earlier).
The performance we get when using hashed sharding is very slightly lower. When a new document is added to the sharded collection, its shard key value needs to be hashed before it's added to the hashed index. However, the overall gain in performance resulting in even distribution could potentially compensate for this loss.
The main disadvantage of using hashed sharding is a decreased number of targeted...
