Parallel processing database system with a shared metadata store
First Claim
Patent Images
1. A parallel processing database system, comprising:
- a master node comprising one or more processors, wherein the master node is configured to receive a query from a client node, the query being communicated to the master node form the client node via one or more networks, develop a query plan based at least in part on the query, determine metadata to be used in connection with execution of the query plan, and to communicate, to a first worker node via at least one of the one or more networks, the query plan and the metadata determined to be used in connection with execution of the query plan, wherein the metadata is stored in a database catalog that comprises database objects, and wherein the metadata is associated with corresponding ones of the database objects;
the first worker node comprising one or more processors, wherein the first worker node is in communication with the master node and receives the query plan from the master node, executes at least a part of the query in connection with executing the query plan received from the master node, and in an event that the first worker node requires additional metadata in connection with executing the at least the part of the query, communicates a request for the additional metadata to a parent node corresponding to the first worker node; and
the parent node comprising one or more processors, wherein in the event that the parent node does not have the additional metadata, the parent node forwards the request for the additional metadata to a next parent node in a hierarchical structure of nodes.
8 Assignments
0 Petitions
Accused Products
Abstract
A method and system for executing database queries in parallel using a shared metadata store. The metadata store may reside on a master node, and the master node may distribute this metadata with query plans to worker nodes. The worker nodes may request additional metadata from the master node when necessary.
-
Citations
24 Claims
-
1. A parallel processing database system, comprising:
-
a master node comprising one or more processors, wherein the master node is configured to receive a query from a client node, the query being communicated to the master node form the client node via one or more networks, develop a query plan based at least in part on the query, determine metadata to be used in connection with execution of the query plan, and to communicate, to a first worker node via at least one of the one or more networks, the query plan and the metadata determined to be used in connection with execution of the query plan, wherein the metadata is stored in a database catalog that comprises database objects, and wherein the metadata is associated with corresponding ones of the database objects; the first worker node comprising one or more processors, wherein the first worker node is in communication with the master node and receives the query plan from the master node, executes at least a part of the query in connection with executing the query plan received from the master node, and in an event that the first worker node requires additional metadata in connection with executing the at least the part of the query, communicates a request for the additional metadata to a parent node corresponding to the first worker node; and the parent node comprising one or more processors, wherein in the event that the parent node does not have the additional metadata, the parent node forwards the request for the additional metadata to a next parent node in a hierarchical structure of nodes. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A method for processing a query, comprising:
-
receiving, by a worker node, a query plan and metadata from a master node via at least one of one or more networks, wherein the query plan is generated by the master node based at least in part on a query that the master node receives from a client node, the query being communicated to the master node form the client node via one or more networks, wherein the metadata is determine metadata to be used in connection with execution of the query plan, and wherein the metadata stored in a database catalog that comprises database objects, and wherein the metadata is associated with corresponding ones of the database objects; executing, by the worker node, at least a part of the query in connection with executing the query plan received from the master node; and in an event that the worker node requires additional metadata in connection with executing the at least the part of the query, communicating a request for the additional metadata to a parent node corresponding to the worker node, wherein in the event that the parent node does not have the additional metadata, the parent node forwards the request for the additional metadata to a next parent node in a hierarchical structure of nodes.
-
Specification