Figure 10.2 shows a typical data distribution across the data warehouse appliance.
Figure 10.2 Data distribution across the data warehouse appliance
From Figure 10.2 we see that data is partitioned and distributed across multiple nodes. In addition to this, typically nodes 1 and 3 will mirror data slices and nodes 2 and 4 will mirror data slices. This brings in several advantages:
· Fault tolerance. Should a node fail, an alternate node can process the query.
· Workload distribution. Should there be a demand for the same data from a lot of users from a querying and analysis perspective, the system ...