To follow on from the original response... it should also be noted that storing large amounts of structured data isn't something that's typically expected or desired in a blockchain. A blockchain like Ethereum accumulates, over time, a copy of every change that's made to the data and since the cost to store and "mine" all of the transactions that change the data gets more expensive (and, for now, slower) as the data accumulates. This has led to efforts such as the integration between Ethereum and IPFS (https://ipfs.io/) where the Ethereum blockchain only stores "pointers" to the data that's versioned in IPFS. Since the pointers are really like addresses that can be stored in a native Ethereum type, they take less space and are easier/quicker/cheaper to mine.
As noted in the original response, you can see some sample components and sample applications that can connect a Pega app to an Ethereum node in the Pega Blockchain Innovation Kit. The documentation in that kit provides some more detail and context for how to query/update an Ethereum code by mapping Pega object data types to matching data types in an Ethereum "smart contract". But be careful about planning to store too much data or complicated data types in Ethereum itself. As you'll in the examples, they are very cautious about what they store in Ethereum.