SPDO Procedure

CLUSTER CREATE Statement

Creates a cluster table.

Requirements:	You must have Control access to all candidate member tables.
	All of the tables must be in the same domain and have the same structure. See Member Table Requirements for Creating Dynamic Cluster Tables in SAS Scalable Performance Data Server: User’s Guide.

Syntax

CLUSTER CREATE cluster-tablename MEMBER=member-name1

[MEMBER=member-name2...][options];

Required Arguments

cluster-tablename

specifies the name of the cluster to be created.

MEMBER=member-name

specifies the name of a member table. To specify multiple tables, use multiple MEMBER= statements.

Alias

MEM=

Optional Arguments

DELETE= YES | NO

specifies whether the cluster and all its members can be permanently deleted with the CLUSTER DESTROY statement. The default is NO.

MAXSLOT= number

specifies the maximum number of slots, or member tables, to be allocated for the cluster. The default setting for the MAXSLOT= parameter is -1. This value configures the server to permit dynamic growth of the number of member tables in a cluster, up to the specified system maximum value. The system maximum value for the number of slots is specified by the MAXGENNUM parameter in the spdsserv.parm configuration file. If there is a known maximum number of slots to be enforced for a particular cluster table, it is more efficient to specify the limitation using the MAXSLOT= argument.

UNIQUEINDEX=YES|NO

specifies whether the unique indexes that are defined in the member tables will be validated and marked as unique in the cluster table. The default setting is YES.

Note: The processing that is required to validate the unique indexes depends on the number of rows in the tables. Processing can take a considerable amount of time for larger tables. If you choose to use the validation process but the indexes are not unique, the CLUSTER CREATE statement fails.

Details

SPD Server supports virtual table structures called cluster tables. A cluster table is a table that consists of numerous similar tables called members. In order to qualify as a member of a cluster table, the tables must share the same table, variable, and index attributes. They must also have the same organization. Cluster tables use the attributes to manage the data that is contained in the members. The cluster table structure provides architecture that enables flexible loading and rapid storage and processing for very large data tables. Cluster tables provide organizational features and performance benefits that traditional SAS tables and regular SPD Server tables do not. For more information about the benefits of and requirements for cluster tables, see “Creating and Using Dynamic Cluster Tables” in SAS Scalable Performance Data Server: User’s Guide.

Example

This CLUSTER CREATE example creates a cluster named Sales_History, which contains member tables sales201601, sales201602, sales201603, sales201604, and sales201606. The cluster table records sales results for the first six months of 2016. The cluster has a limit of 24 member tables.

proc spdo library=libref;
cluster create Sales_History
mem=sales201601
mem=sales201602
mem=sales201603
mem=sales201604
mem=sales201605
mem=sales201606;
maxslot=24;

Last updated: February 3, 2017