Hashing a block

In the PoW systems, new entries to a blockchain require hashes to be computed. In Bitcoin, miners must compute two SHA-256 hashes on the current transactions in the block—and included therein is the hash of the previous block.

This is pretty straightforward for a hashing algorithm. Let’s briefly reiterate: an ideal hash function takes the expected input and then outputs a unique hash. It is deterministic. There is only one possible output and it is impossible (or computationally improbable) to achieve that output with a different input. These properties ensure that miners can process a block and that each miner can return the same result. It is through hashing that Blockchains attain two properties that are crucial to their adoption and current popularity: decentralization and immutability. 

Linking the current block to the previous block and the subsequent block is in part what makes the blockchain an ever-growing linked list of transactions (providing it with the property of immutability), and the deterministic nature of the hash algorithm makes it possible for each node to get the same result without issue (providing it with decentralization).