This topic is all about the implementation of Blockchain in enterprises. In the first part, we will take a general look at the integration of external data. In the second part, we will present challenges to run a blockchain network. All data in a blockchain is stored as transactions. Transactions should not contain too much information, because this would enlarge the distributed ledger too much. The transaction size of most of the presented use cases is small. Identity information and sums of money are stored in transactions for the foreign currency exchange use case. For the Smart Grid use case, transactions include the amount and price of the traded electrical power as well as the identities of the producer and the consumer. However, the Blockchain for Education use case needs to store digital certificates that could be quite large. For this use case, not the digital certificates itself but their fingerprints are stored in the transactions. Fingerprints are unique bit strings of a certain length, which can be computed from any digital information such as text, documents, videos or other data sets. Blockchain infrastructures provide asymmetric cryptographic hash functions, which can be used to convert any digital information into a unique bit string. The bit strings – also called hash values – are unique, because even the slightest changes in the digital information makes a completely different bit string. In addition, it is impossible to determine the original digital information from the fingerprint. Besides the property of being small, fingerprints can be used to retrieve the original digital information and to verify the integrity of the corresponding digital information. A blockchain cannot access any data outside its network; however, fingerprints are an important mean to refer to external data that is stored in databases, cloud storage or other repositories. For example, the digital certificates used in the Blockchain for Education use case are stored in a document management system. Fingerprints of documents could also be applied in application areas such as proof of origin, supply chain, medical engineering or the media industry whenever immutable documentation is necessary. In production processes, operating conditions of machines, compliance with process regulations or the origin of products are also documented in external repositories. If such data is stored as fingerprints in transactions on the blockchain, then the data cannot be manipulated, which could enable real-time audits or simplify auditing by external auditors. A second group of external data are sensor values. Especially in the application areas Internet of Things and supply chain, sensor technology in combination with smart contracts is used to automate processes. In the food chain use case the sensors support food traceability, raise alarms and trigger actions. The sensors deliver values that are external to the blockchain. Therefore, it is important to check the identity of the sensors and the sensor values before they are used and stored on the blockchain. Similar correctness issues occur, when online information is used in blockchain-based processes. These issues have led to the development of oracles that are third party services, which find and verify sensor values or online information. In summary, integration of data stored in application systems is important for the automation of existing processes in enterprises using blockchain technology. Fingerprints can be used to point to any data set in application systems and can be used to verify the integrity of the data set. Oracles are used to validate sensor values and online information.
