Dell DR6300 Microsoft DPM - Setting Up the DR Series System as a VTL Backup Ta - Page 48

B Managing VTL media and space usage, B.1 General performance guidelines for DMA configuration

Page 48 highlights

B B.1 Managing VTL media and space usage General performance guidelines for DMA configuration • The DR Series system (version 3.2 and later) provides inline VTL deduplication, compression, and encryption at rest functionality. Backup applications (such as Dell NetVault, Symantec BackupExec, Symantec NetBackup, and so on) should be configured so that any multiplexing, pre-compression, software-side deduplication, or encryption is disabled. Enabling any of these features may adversely affect the space savings and ingest performance of the DR Series system VTL feature. • Slots and media should be configured so as to accommodate the environment backup requirements. Initially, the logical capacity of a VTL should be no more than twice the physical size of the DR Series system. If the initial VTL setup is over-subscribed at higher than a 2-1 ratio without proper planning the DR Series system could fill up prematurely and cause unexpected system outage. It is highly advisable to configure the DR Series system VTL feature such that the media count be made to accommodate your initial data protection requirements. and then media be added as the deduplication statistics become available to ascertain growth, media, and space requirements. • Media Type selection will depend on a number of factors including the DMA used, the backup cycles, data sources, and more. As a general rule, using smaller tapes is better than using larger tapes so as to allow for a higher level of control over space usage by backup operations. This also allows for easier handling in the event of a system running out of physical space as well as the normal data cleanup procedures. • Adding media to an existing DR Series system VTL is painless and should be leveraged to incrementally add media as needed. Although this may require a higher level of involvement in managing the media usage, it will result in better performance and avoid unplanned outages. B.2 VTL space sizing and planning Various factors such as total data footprint, change rate, backup frequency and data lifecycle policies will dictate how much physical space will be needed to accommodate the Virtual Tape Libraries within a DR Series environment. In addition, if other container types are hosted these two must be factored into space requirement calculations. As a general rule the following can be used as a reference architecture to determine the basic capacity needed for a given virtual tape library container: 1. Determine Existing Data Set 2. Determine the change rate (Differential) 3. Determine the retention period 4. Calculate the data footprint during the retention period for existing data sets based on a 10-1 deduplication ratio 48 Setting Up the Dell™ DR Series System as a VTL Backup Target on Microsoft® Data Protection Manager | January 2016

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52

48
Setting Up the Dell™ DR Series System as a VTL Backup Target on Microsoft® Data Protection Manager | January
2016
B
Managing VTL media and space usage
B.1
General performance guidelines for DMA configuration
The DR Series system (version 3.2 and later) provides inline VTL deduplication, compression, and
encryption at rest functionality. Backup applications (such as Dell NetVault, Symantec BackupExec,
Symantec NetBackup, and so on) should be configured so that any multiplexing, pre-compression,
software-side deduplication, or encryption is disabled. Enabling any of these features may adversely
affect the space savings and ingest performance of the DR Series system VTL feature.
Slots and media should be configured so as to accommodate the environment backup requirements.
Initially, the logical capacity of a VTL should be no more than twice the physical size of the DR Series
system. If the initial VTL setup is over-subscribed at higher than a 2-1 ratio without proper planning
the DR Series system could fill up prematurely and cause unexpected system outage. It is highly
advisable to configure the DR Series system VTL feature such that the media count be made to
accommodate your initial data protection requirements. and then media be added as the
deduplication statistics become available to ascertain growth, media, and space requirements.
Media Type selection will depend on a number of factors including the DMA used, the backup cycles,
data sources, and more. As a general rule, using smaller tapes is better than using larger tapes so as to
allow for a higher level of control over space usage by backup operations. This also allows for easier
handling in the event of a system running out of physical space as well as the normal data cleanup
procedures.
Adding media to an existing DR Series system VTL is painless and should be leveraged to incrementally
add media as needed. Although this may require a higher level of involvement in managing the media
usage, it will result in better performance and avoid unplanned outages.
B.2
VTL space sizing and planning
Various factors such as total data footprint, change rate, backup frequency and data lifecycle policies will
dictate how much physical space will be needed to accommodate the Virtual Tape Libraries within a DR
Series environment. In addition, if other container types are hosted these two must be factored into space
requirement calculations. As a general rule the following can be used as a reference architecture to
determine the basic capacity needed for a given virtual tape library container:
1.
Determine Existing Data Set
2.
Determine the change rate (Differential)
3.
Determine the retention period
4.
Calculate the data footprint during the retention period for existing data sets based on a 10-1
deduplication ratio