How are transactions on large data done in a smart contract?

Question

I am presently working on a project involving read and write operations on a huge database. The database is structured in such a way that there is a key, and for the key there is a corresponding set of data. Now, I would like to consider each of these operations as transactions in a smart contract. There are two ways of doing this:

1. Construct a struct that would contain the key as well as the set of data corresponding to the key.

2. Simply store the keys and a pointer to the set of data stored in an external database. I was thinking of using Oraclize for accessing the database.

If the first option is considered, it is definitely not a practical choice as storing such large data directly on a blockchain is not possible. But if the second option is considered, then isn't the external database prone to tampering, thus having no point to build it on a blockchain.

Effectively, I am looking for building a blockchain based application that deals with Big data keeping the security of the data in mind.

Answer 1

Indeed - if the data is not stored and structured properly - it is prone to tampering.

One of the solutions is to store a cryptographic hash in the blockchain and structure your data in such a way that it is possible to reliably calculate the hash to verify it has not been tampered.

You can do it in many different ways - in the end it is going to be something resembling a Merkele tree. IPFS is the most straightforward option - the data is kept in such a way that it is identified by its cryptographic hash that you store in the Ethereum contract.