Networker supports all sorts of devices such as tape, file type, advanced file type, cloud and Data Domain for storage. These can be either connected to the Networker server (local devices) or network-attached (remote) devices.
File Type Devices are backward-compatible disk storage. They can be local or NFS only (not CIFS), and they can accommodate only one backup stream.
Advanced File Type Devices can be local, remote, NFS, CIFS. Multiple backup streams can be written to AFTD.
DD Boost devices are for use with Data Domain, and Cloud devices can be written to as in teh case of EMC Atmos.
Client direct backup is when there is a CIFS, NFS or DD device mounted to the storage node as storage. The backup stream can bypass the storage node and write directly to the storage share. It is enabled by default, and Windows clients are not supported for *NIX storage nodes (and vice-verse)
Dedicated Tape Libraries are configured as one tape library per storage node. Shared Tape Libraries can be static, where each drive is dedicated to a specific storage node (often used wiht VTL) or Dynamic, where at least one drive is shared among multiple nodes - typicall used with a SAN.
Multiplexing backup streams is a technique to maximize throughput of the tape by interleaving bacup streams. This is determined by two parameters: server parallelism and device target sessions.
Open Tape Format (OTF) is a heterogeneous storing of multiplexed data that can be read by any hardware vendor supporting the format. This is for future-proofing the backup save sets from proprietary protocols.
Pools are collection of volumes and used to determine where save sets will be written. A separate pool must be created for clone sets, because they can not mix with regular backup data. The selection criteria for how data is written to a pool is: Group-client-save set-level.
Volumes must be labelled to be used in a pool, and they can be labelled manually or through the use of a template upon addition to the system.
Clones are copies of data from one volume to another. Use NMC or nsrclone to initiate a clone operation. Clones are used typically for DR purposes, and can take place between 2 disparate devices. The destination volume must reside in a clone pool, and you can use filters to select data to be cloned. Scheduled clone jobs are available after Networker 7.61 SP1 as well.
Staging is moving data from one volume to another and can be between 2 disparate devices or media types. After the save set is staged, the media database is updated with the new location. Save sets can be staged manually or automatically based on the number of days the data has resided on a given pool, or based on a high/low watermark.
No comments:
Post a Comment