Grakn is a distributed hyper-relational database for knowledge-oriented systems. Grakn enables machines to manage complex data that serves as a knowledge base for cognitive/AI systems.
Graql is Grakn's reasoning (through OLTP) and analytics (through OLAP) query language. Graql is a much higher level abstraction over traditional query language - SQL, NoSQL, or Graphs.
Cassandra is highly available system, uses P2P, AP in brewers theorem, multi data center support, no overhead of moving OLTP data to OLAP if considering dse.
HBase is CP system, tight integration with Hadoop
Иметь ввиду
Возможно, как способ хранения и мне подойдёт
Из видео https://youtu.be/OeFrudRlXAM?t=17m8s на странице https://grakn.ai/pages/overview/index.html увидел, что оно построено поверх Апачевских Cassandra, HBase (хранилища), Hadoop, Spark (кластеризация), Kafka (брокер сообщений (использован для индексации)) и Tinkerpop (обработка графов)
в https://www.quora.com/Why-would-someone-use-Cassandra-over-HBase-Are-they-both-similar-or-does-one-solve-a-specific-type-of-problem-better-than-the-other написано про первые три из них: