【正文】 to a disk group ? Disk devices can be added one or more at a time ? Disk devices are ungrouped from a disk group ? You can only do this one disk at a time ? Disk devices within a disk group fail (assuming VRAID survival) ? Leveling is dynamic (there is no user control) ? Leveling is detectable through Command View EVA ? Customer notice CN0496W on SAW has more information on leveling 1 – 17 Virtual disk leveling ? Virtual disk blocks are automatically relocated to level spindle use Moderately redundant volume (VRAID5) No redundancy volume (VRAID0) Highly redundant volume (VRAID1) + Add more disks Additional storage space available No redundancy volume (VRAID0) Highly redundant volume (VRAID1) Moderately redundant volume (VRAID5) Available storage space (equivalent of 3 disks) VRAID1 pair VRAID1 pair VRAID1 pair VRAID1 pair VRAID1 pair VRAID1 pair VRAID1 pair VRAID1 pair VRAID1 pair VRAID1 pair 1 – 17 Distributed sparing ?The allocation of space per disk group to recover from physical disk group member failure ?Disk failure protection (spare) space is allocated across all members of a disk group ?You assign the amount of protection capacity you want to reserve ? When you create the disk group ? After the disk group has been created ?Disk failure protection level (assuming samesized drives) ? None ? Single — Requires space of two disk drives per disk group ? Double (in most instances, not concurrent failures) — Requires the space of four disk drives per disk group 1 – 20 Rev. – HP Restricted x Distributed sparing ? Redundancy automatically regenerated (1st priority) ? Data redistributed across the virtual pool (2nd priority) Available storage space (virtual space = 2 disks) Moderate redundant volume (VRAID5) High redundant volume (VRAID1) Redundancy temporarily promised x Moderate redundant volume (VRAID5) High redundant volume (VRAID1) Available storage space (virtual space 1 disk) 1 – 21 Reconstruction ?After an original volume bees inaccessible, this is the process by which data is regenerated and moved (migrated) to other storage in the disk group ?Occurs after disk failure ? Userwritten virtual disk blocks are reconstructed on remaining members of RSS and disk group ? Disk group is leveled ? All VRAID0 virtual disks are inoperative ?Always allow reconstruction to plete ?As soon as reconstruction has pleted, replace each failed disk ?Controller event log indicates Missing ? Reconstructing 1 – 28 Rev. – HP Restricted Hosts and LUNs ? A host is a collection of host bus adapters that reside in the same (virtual) server ? A LUN is a virtual disk presented to one or multiple hosts ? Rules for hosts, connections, and LUNs ? A maximum of 256 hosts ? A maximum of 1,024 Fibre Channel Adapter connections ? A maximum of 256 LUNs on any one Fibre Channel Adapter ? A maximum of 8,192 presentations of LUNs to hosts ? Examples: ? 1 LUN presented to 1 host = 1 presentation ? 1 LUN presented to 256 hosts = 256 presentations ? 256 LUNs presented to 1 host each = 256 presentations 1 – 29 Rev. – HP Restricted HSV2x0 controller identification by host port ? Last character of the port WWN ? Controller A ? Port 1: 8 ? Port 2: 9 ? Port 3, A ? Port 4: B ? Controller B ? Port 1: C ? Port 2: D ? Port 3, E ? Port 4: F 8 9 A B FP1 FP2 MP2 DP1B DP2B DP1A DP2A MP1 FP3 FP4 C D E F HSV210 Controller A Controller B C D FP1 FP2 MP2 DP1B DP1A 8 9 MP1 HSV200 Controller A Controller B 14 – 1819 Rev. – HP Restricted HSV300 controller identification by host port ? Last character of the port WWN ? Controller 1, host port 1: 8 ? Controller 1, host port 2: 9 ? Controller 2, host port 1: C ? Controller 2, host port 2: D C D 8 9 Controller 1 Controller 2 FP1 FP2 DP1B DP1A 14 – 19 Rev. – HP Restricted HSV4x0 controller identification by host port ? Uses same numbering as HSV210 ? Controller A ? Port 1: 8 ? Port 2: 9 ? Port 3: A ? Port 4: B ? Controller B ? Port 1: C ? Port 2: D ? Port 3: E ? Port 4: F FP1 FP2 MP2 DP1A DP2A MP1 FP3 FP4 DP3A DP1B DP2B DP3B FP1 FP2 MP2 DP1A DP2A DP1B DP2B MP1 FP3 FP4 HSV400 HSV450 14 – 20 SSSU description ?What is the SSSU? ? Command line application that allows configuration and control of the controllers and storage system ? Allows plex and repetitious configurations to be scripted and executed ? Allows you more flexibility than Command View EVA ?Configuration ? SSSU is installed on management server with the Command View EVA master installer or on a host with access to Command View EVA ? Communicates using TCP/IP with Command View EVA running on the management server 15 – 2 Terminology (1 of 2) ? System — Controller or controller pair that is treated conceptually as one controller. In Command View EVA, this is the name you give to the initialized storage system. ? Disk — Physical storage device (disk drive) ? Disk group — Collection of disks from which storage is created ? Folder — A logical anization of objects ? Host — A system that has data path access to the virtual disk through a LUN defined on the system ? LUN — Hostaccessible presentation of storage on the system ? Manager — Command View EVA management server that controls configuration of the storage system 15 – 3 Rev. – HP Restricted Terminology (2 of 2) ?Controller — Controller enclosure ?Storage — A virtual disk that may may be configured as VRAID0, VRAID1, VRAID5, or VRAID6 for data storage ?Spare policy — Disk failure protection level (None, Single, and Double) 15 – 3 Rev. – HP Restricted Common configuration mands ?SELECT — Selects Command Vi