To reduce the database entry size, the existing database entry format is changed. The internal representation of an entry changed from an ASCII LDIF format to a tagged binary format. The data stored in the database does not have the characteristic starting of dn: anymore, the first byte of an entry being a value bigger than 0xE0 (hence all values 0xE0 to 0xFF are to be considered reserved for internal use).
For compatibility reasons entries can be a mix of LDIF and binary representations, but any modification will write the entry in binary format.
Suffix entries data can be compressed when written to disk to minimize their disk footprint. Compression is enabled according to the settings of the compression-mode and compression-entries properties,
For additional information, refer to the Chapter 8, Writing Entry Store and Entry Fetch Plug-Ins, in Sun Directory Server Enterprise Edition 7.0 Developer’s Guide.
To save disk space, you can restore a server by moving files in place of copying them. You can perform the copyless restore by setting a flag with the restore command.
For more information, see Binary Restore in Sun Directory Server Enterprise Edition 7.0 Administration Guide
Instances installed on Windows systems now support Internet Protocol version 6.
Likewise, server instances on other operating systems also support IPv6.
The dsutil command now performs the functions formerly provided by the ns-activate, ns-inactivate, and ns-accountstatus commands.
Backup operations perform a database verify on archived data when the --flags verify-db option is specified.
The index filter analyzer identifies index lists where the number of entries exceeds the maximum number of indexable entries (the ALLID threshold) and monitors user searches using such index lists. The index filter analyzer is enabled using the dsconf enable-index-filter-analyzer command.
Entry aggregation results in the following:
Optimizes queries to secondary data view
Searches for secondary data view first when required
Handles large result sets (VLV control) better
Requests Grouping to secondary source
JDBC data view now supports Date and Blob.
Directory Proxy Server now uses a new logging engine implementation that performs more efficiently on multi-core systems.
New criteria based on LDAP groups
Management of the maximum throughput
New type of data view to address more use cases, for example, company mergers.
An enhanced regex distribution algorithm is added, as described in Configuring Pattern Matching Distribution Algorithm in Sun Directory Server Enterprise Edition 7.0 Administration Guide.
To optimize the performance of searches of a join data view, Directory Proxy Server makes use of Virtual List View (vlv) indexes. It helps you to avoid the scenario where search hits the size limits due to the lots of entries from one data source and very few from the others. To use VLV indexes, see Browsing Index in Sun Directory Server Enterprise Edition 7.0 Reference.