Data Replication

It is possible to specify to what nodes data should be replicated, that is, replication strategies are configurable. A common approach is to have a complete copy of all data in each Data Center. That is, a complete copy of all data in every geographical location. Internally in a specific Data Center, it is possible to specify on how many nodes a copy of a specific data should be stored. This is specified by the Replication Factor. With a Replication Factor of two, the data will be stored on two different system database nodes within the Data Center.

In a cluster like in figure 2, it is a good idea to have a Replication Factor of 2 for DC-1 and DC-2. With Replication Factor 2, the Data Center has internal redundancy of the data. DC-3 only has one system database node, so the Replication Factor will be one.

The total number of data copies that are replicated is referred to as the replication factor. A replication factor of 1 means that there is only one copy of each row in a Data Center, whereas a replication factor of 3 means three copies of the data are stored across the Data Center.

For ASP 113 data is always replicated to all database nodes in a Data Center. The replication factor for a Data Center must be equal to the number of Cassandra servers in the Data Center.

Data load sharing is not used in ASP 113.