HP ProLiant DL380G5-WSS 3.7.0 HP StorageWorks HP Scalable NAS File Serving Sof - Page 140

For example, if there is no backlog (B=1) and 1000 files change per node per interval on a 16-node cluster, the formula would be: S = (4KB * 1000) * 17 = 68000KB = 68MB If 100,000 files changed, the size would be 6.8GB. Similarly, if 750,000 files changed, the size would be 51GB. Add 1GB to the total if the Performance Dashboard will be used and 1GB if the mxfs reply cache will be configured on the administrative filesystem. Create the filesystem The administrative filesystem must be created manually. Before creating the filesystem, obtain the UUIDs and partition numbers for the volumes or disks that will be used for the filesystem. The disk or disks used for the administrative filesystem do not need to be imported into the cluster and can be either partitioned or unpartitioned. IMPORTANT: The administrative filesystem must be created on its own LUN. If replication is used, a high-performance LUN such as a RAID1 or VRAID1 LUN should be used. The sandiskinfo -a command lists the disk UUIDs and partitions for all disks imported into the cluster. To see this information for disks/LUNs that have been not been imported into the cluster, add the -u option to the command. When you have obtained the UUIDs and partition numbers, run the following command to create the administrative filesystem: mx config adminfs set [--reuse] [--path ] You can specify one or more partitions (or unpartitioned disks). For partitioned disks, is a list of the partitions to be used for the filesystem. Specify each partition as , where is the uuid of the disk and is the partition number on the disk. If the disk is not partitioned, specify only the or enter 0 as the partition number. The arguments are: [--path ] By default, the administrative filesystem is mounted at /_adminfs . To mount the filesystem at another location, use this option to specify the mount point. [--reuse] 140 Configure PSFS filesystems