Storing queued documents on disk

Documents in queued input/output jobs are by default stored as blobs (Binary Large Objects) in the runtime repository. You can, for each queue, select to store the blobs as files on disk instead. The runtime repository will in this case include references to the blobs stored on disk.

There are several reasons to store blobs on disk, for example to:

•	Improve performance.

•	Decrease the amount of runtime repository data to backup.

•	Minimize the risk to reach maximum database and tablespace limits.

Storing blobs on a network share

To store blobs on a network share, you must use UNC (Uniform Naming Convention) when you specify the path. You cannot use drive letters.

For example, you can use \\wtvs01\data\queues but not H:\data\queues when you specify the path.

When not to store blobs on disk

•	If you intend to share the queue between several StreamServer applications (load balancing and fail over), you cannot store blobs on disk.

•	StreamStudio Reporter can only view the status of the jobs in a queue if the blobs are stored on disk. This means StreamStudio Reporter will not be able to resend or delete jobs in this case.

•	StreamStudio Collector cannot be used if you select to store blobs on disk. Note that archiving must be disabled for the output connector if you select to store blobs on disk.

To store blobs on disk

1	Activate the generic layer.

2	Right-click the Platform view and select Manage Queues. The Manage Queues dialog box opens.

3	Select the queue to configure.

4	Select the Advanced tab.

5	Select Store blobs on disk.

6	In Blob path, specify the path to the folder where to store the files.