In this chapter, we discovered various features of Cassandra that make it an ultrafast, highly available, distributed database with no single point of failure. We discussed how these features are achieved by Cassandra internally, for example, we discussed how data is distributed and then replicated among nodes of a Cassandra cluster. We discussed various replication strategies that are available with Cassandra and the criteria to choose one. We also discussed various data distribution algorithms available with Cassandra. We saw how the vnodes feature in Cassandra enables more efficient data distribution and load sharing among nodes. In the next chapters, we'll discuss in detail how Cassandra does a read/write operation internally and the data structures it has to server a read/write request. We'll also discuss the various techniques it uses to persist data and the uses of various techniques such as caching or compaction to make our queries faster.