Specifies the bucket size for a new data set
| Language element: |
data set option
|
| Default: |
32
|
| Engines: |
CONCUR
|
| Valid as: |
data set option, engine or host option
|
| OpenVMS specifics: |
All aspects are host-specific
|
-
bucket-size
-
can range from 0 to 63. If you specify a value of 0, SAS uses the minimum number of blocks that are needed to contain a single
observation.
The BKS= data set option specifies the number of OpenVMS disk blocks in each bucket of a
new data set. A disk block
is 512 bytes. A bucket
is a storage structure of a set number of blocks used for building
and processing files of relative and indexed organization. A bucket contains one or more records or record cells. Buckets are the unit of contiguous transfer between
RMS buffers and the disk.
The BKS= data set option is applied only when the data set is created. When deciding on the bucket size to use, consider whether the file will be
accessed randomly (small bucket size), sequentially (large bucket size), or both ways (medium bucket size). For example, if you know that all the files in a SAS data library will be accessed randomly,
by using the FSEDIT procedure for example, then use a small bucket size at the creation of each file to optimize
the access. The following example uses the BKS= option in the LIBNAME statement so that the bucket size applies to all
CONCUR data sets in that library:
libname test concur '[randomdir]' bks=30;
The BKS= data set option corresponds to the FAB$B_BKS field in
OpenVMS RMS or to the FILE BUCKET_SIZE attribute when you use File Definition Language (FDL). For additional details, see
Guide to OpenVMS File
Applications.
Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.
