Loading...

Indexing Version 2: Overview

SnapProtect now provides a next-generation version of Indexing (called Indexing V2). It has a new internal architecture that supports some changes to the configuration and operation of Indexing. Although Indexing V2 changes the way backup and indexing are accomplished within the MediaAgent tier, the basic information flow for indexing remains the same (see Index: Overview).

Indexing V2 is included in this release without installing any additional software or licenses. Version 2 is only available for fresh installations of clients; clients already using Indexing Version 1 continue to use Version 1 and cannot be migrated to Version 2. For a list of agents that supporting Indexing Version 2, see iDataAgents that Use SnapProtect Indexing.

Changes from Version 1

A few key changes for Version 2 are the use of transaction logs for generating indexes, implementation of a single cumulative index across cycles for backup sets, and the option to have a dedicated Indexing MediaAgent.

Indexing MediaAgent

Index creation is now logically separated from the data backup process. This allows locating backup and indexing functions on separate hosts, though they can still be collocated if appropriate for a given installation. It also removes index creation as a dependency for completing a data backup.

Transaction Logs

The index is constructed on the Indexing MediaAgent from transaction logs that the Backup MediaAgent creates while backing up data. Creating these logs takes less processing and time than creating the actual index database, and is an inherently less error-prone process. This approach insulates backup progress from index processing time or issues.

Cumulative Index

In V1, each subclient had its own complete, independent index for each cycle, which was backed up after every backup operation. In V2, a single index contains the metadata for all subclients in a backup set, for all backup operations. This single index for each backup set, located in the Index Directory, substantially reduces the backup media footprint.

Advantages of Version 2

Indexing Version 2 provides these advantages over Indexing Version 1.

Reduced Library Space Consumption

The new cumulative index (see above) is not backed up at the end of each data backup operation. The transaction logs are backed up at the end of the job. The index is only backed up when index backup criteria are met.

GridStor

When GridStor is deployed, Indexing Version 2 removes the time and disk space required to copy an index between MediaAgents if the backup process fails over to a different MediaAgent.

Browse Deleted Items

With Indexing Version 2, browse operations show deleted items across backup cycles.