Dell DR4300 DR Series System Administrator Guide - Page 110

Dell DR Series System, Interoperability Guide, Direct I/O memory

Page 110 highlights

The Rapid NFS plugin (driver) must be installed on client systems After the plug-in is installed, write operations will go through Rapid NFS while metadata operations such as file creates and permission changes will go through the standard NFS protocol. Rapid NFS can be disabled by uninstalling the plug-in. Markers must be set on the client, not in the DR Series GUI If you are using a DMA that supports a marker, should explicitly set it. Your containers should have the marker type of None until you set the marker using the Mount command on the client (after installing the Rapid NFS plug-in). For existing containers, re-set the marker using the procedure that follows. For example, if you wanted to set the CommVault marker (cv): mount -t rdnfs 10.222.322.190:/containers/backup /mnt/backup -o marker=cv Mount command usage: rdnfs [nfs mount point] [roach mount point] -o marker=[marker] where: nfs mount point = Already mounted nfs mountpoint roach mount point = A new mount point marker = appassure, arcserve, auto, cv, dump, hdm, hpdp, nw, or tsm Your DR Series Rapid NFS is available on a DR Series system and a client with a minimum of 4 CPU cores system must meet the running at a minimum of 4 GHz cumulative processing power and 2 GB memory. Kernels must minimum be 2.6.14 or later. For a list of supported operating systems, see the Dell DR Series System configuration Interoperability Guide. If you update your operating system, you must update your Rapid NFS plug-in as well. Updates are available on the Support site as well as within the GUI on the Clients page. Rapid NFS is stateful If the DR Series system goes down, the connection will terminate. DMAs will restart from the last checkpoint. Rapid NFS and passthrough mode If Rapid NFS mode fails for any reason, the DR Series system falls back to regular NFS mode automatically. For details, see Monitoring Performance. Rapid NFS performance considerations When using Rapid NFS on your client, Dell recommends that you do not run other protocols to the DR in parallel, as this will adversely affect your overall performance. Rapid NFS acceleration constraints Rapid NFS does not support: • Direct I/O memory • Mapped files • File path size greater than 4096 characters • File write locks across clients NOTE: If the client and server do not have the same times, the times seen will not match typical NFS behavior due to the nature of file system in user space (FUSE). 110

  • 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
  • 53
  • 54
  • 55
  • 56
  • 57
  • 58
  • 59
  • 60
  • 61
  • 62
  • 63
  • 64
  • 65
  • 66
  • 67
  • 68
  • 69
  • 70
  • 71
  • 72
  • 73
  • 74
  • 75
  • 76
  • 77
  • 78
  • 79
  • 80
  • 81
  • 82
  • 83
  • 84
  • 85
  • 86
  • 87
  • 88
  • 89
  • 90
  • 91
  • 92
  • 93
  • 94
  • 95
  • 96
  • 97
  • 98
  • 99
  • 100
  • 101
  • 102
  • 103
  • 104
  • 105
  • 106
  • 107
  • 108
  • 109
  • 110
  • 111
  • 112
  • 113
  • 114
  • 115
  • 116
  • 117
  • 118
  • 119
  • 120
  • 121
  • 122
  • 123
  • 124
  • 125
  • 126
  • 127
  • 128
  • 129
  • 130
  • 131
  • 132
  • 133
  • 134
  • 135
  • 136
  • 137
  • 138
  • 139
  • 140
  • 141
  • 142
  • 143
  • 144
  • 145
  • 146
  • 147
  • 148
  • 149
  • 150
  • 151
  • 152
  • 153
  • 154
  • 155
  • 156
  • 157
  • 158
  • 159
  • 160
  • 161
  • 162
  • 163
  • 164
  • 165
  • 166
  • 167
  • 168
  • 169
  • 170
  • 171
  • 172
  • 173
  • 174
  • 175
  • 176
  • 177
  • 178
  • 179
  • 180
  • 181
  • 182
  • 183
  • 184
  • 185
  • 186

The Rapid NFS plug-
in (driver) must be
installed on client
systems
After the plug-in is installed, write operations will go through Rapid NFS while metadata
operations such as file creates and permission changes will go through the standard NFS
protocol. Rapid NFS can be disabled by uninstalling the plug-in.
Markers must be set
on the client, not in
the DR Series GUI
If you are using a DMA that supports a marker, should explicitly set it. Your containers should
have the marker type of
None
until you set the marker using the Mount command on the client
(after installing the Rapid NFS plug-in). For existing containers, re-set the marker using the
procedure that follows.
For example, if you wanted to set the CommVault marker (cv):
mount -t rdnfs 10.222.322.190:/containers/backup /mnt/backup -o
marker=cv
Mount command usage:
rdnfs [nfs mount point] [roach mount point] -o marker=[marker]
where:
nfs mount point = Already mounted nfs mountpoint
roach mount point = A new mount point
marker = appassure, arcserve, auto, cv, dump, hdm, hpdp, nw, or
tsm
Your DR Series
system must meet the
minimum
configuration
Rapid NFS is available on a DR Series system and a client with a minimum of 4 CPU cores
running at a minimum of 4 GHz cumulative processing power and 2 GB memory. Kernels must
be 2.6.14 or later. For a list of supported operating systems, see the
Dell DR Series System
Interoperability Guide
.
If you update your operating system, you must update your Rapid NFS plug-in as well. Updates
are available on the Support site as well as within the GUI on the
Clients
page.
Rapid NFS is stateful
If the DR Series system goes down, the connection will terminate. DMAs will restart from the
last checkpoint.
Rapid NFS and
passthrough mode
If Rapid NFS mode fails for any reason, the DR Series system falls back to regular NFS mode
automatically. For details, see
Monitoring Performance
.
Rapid NFS
performance
considerations
When using Rapid NFS on your client, Dell recommends that you do not run other protocols to
the DR in parallel, as this will adversely affect your overall performance.
Rapid NFS
acceleration
constraints
Rapid NFS does not support:
Direct I/O memory
Mapped files
File path size greater than 4096 characters
File write locks across clients
NOTE:
If the client and server do not have the same times, the times seen will not match
typical NFS behavior due to the nature of file system in user space (FUSE).
110