HP vertica the physical architecture is designed to distribute physical storage and to allow parallel query execution over a potentially large collection of computing resources.
The most important terms to understand are instance, node, cluster, and data:
NODE: a node is a machine configure to run vertica
Instance: Its the vertica instance installed on the node
Cluster: A cluster consists of collection of nodes
data: Data is the data in the vertica database distributed across the nodes
Vertica platform is build ground up. there are 6 important vertica features:
Columnar Orientation: Data is stored in columns instead of rows, thus saving memory and I /O
Advance Compression:Operates on compressed and encoded data delivers superior performance on a smaller storage footprint and reduced capital expense
High availability: Vertica is designed for high availability. to achieve this vertica uses projections or copy of projections stored in different nodes. number of projections are based on K-Safety=1
K-Safety is the measure of fault tolerance of the database. it support values 0,1,2. inorder to enable K-Safety there should be minimum of 3 nodes in the cluster
Automatic Database design: The database designer allows you to optimize, tune and control your database without shutting down the database
Application integration: Using standard SQL, there is seamless integration with robust and ever growing analytical solutions
Massively Parallel Processing:automatic,limitless, linear scaling of peer to peer nodes, each processing an allocated subset of data
No comments:
Post a Comment