Thin Clones
Thin clones are clone volumes that share space with their parent volumes, which are referred to as template volumes. When first created, a thin clone shares the same data as its template volume. Thin clones are counted towards the total number of volumes in the group.
Volume Administrator Accounts
Beginning with version 5.0, array firmware supports the creation of volume administrator accounts. Volume administrator accounts are created by group administrators and may then be assigned a quota of storage to manage within one or more pools. They can create and manage volumes within their quota, and can perform all operations (including replication to designated partners) on volumes they own.
Volume administrators can view information only for pools and volumes to which they have access. For security purposes, the volume administrator has a limited view of group and pool configuration settings, and cannot view configuration settings tor events that might enable them to gain additional access.
The creation of volume administration accounts enables group managers to delegate the management of volumes to other administrators without exposing sensitive group security information.
Tiered HDD/SSD Configurations with RAID-6 Accelerated
With the V5.0 firmware, support for two new products is being introduced. The PS6000XVS and PS6010XVS offer tiered configurations in which solid-state disks (SSD) and serial-attached SCSI (SAS) drives reside in the same chassis. These products are optimized to provide SSD level performance and hard disk drive (HDD) level capacity for tiered applications.
A new RAID policy, called RAID-6 Accelerated, is available on XVS array models. RAID-6 Accelerated provides RAID 6 parity protection, while optimizing the use of solid-state drives for optimal performance.
In XVS systems, one hard disk drive is configured as a spare and provides redundancy protection in the event of either an SSD or HDD failure. In the event of an SSD failure, the RAID set is reconstructed using the HDD spare. During this time, array performance is temporarily degraded. When the failed SSD is replaced, the data is copied back to the new SSD, the HDD returns to its status as a spare, and performance returns to optimal levels.
Note:Maintaining an on-site spare for SSD drives used with XVS systems will shorten the period of performance degradation caused by using the HDD spare in the SSD RAID set.
You can configure these types of arrays using the Remote Setup Wizard included with version 3.3.2 or higher of the Host Integration Toolkit.
Drive Spin-Down
To save energy, spare drives in an array will be spun down (SAS) or put into sleep mode (SATA). The drives will be automatically spun up or awakened for internal scans and when needed to replace a failed drive in a RAID set. This feature applies only when there are no more than two spare drives in an array. After you upgrade all group members to PS Series Firmware Version 5.0, this feature will be available on the following array models: PS6000 (all suffixes), PS6500 (all suffixes), PS6010 (all suffixes), and PS6510 (all suffixes).
Lifting of Pool Membership Restrictions for PS5500, PS6500, and PS6510 Arrays
As of the PS Series Firmware Version 5.0 release, there is no more requirement that PS5500, PS6500, and PS6510 array models (4U, 48-drive chassis) must be placed in a pool which is separate from other array models. After you upgrade all group members to PS Series Firmware Version 5.0, you can move up to eight array members to any pool in the group.
Note:Volume and page metrics may appear skewed in favor of the larger arrays in a pool. Because of this, and for performance reasons, Dell does not recommend mixing different array models in the same pool in most environments.
Increased Capacity, Replication, and Connection Limits
Beginning with PS Series Firmware Version 5.0, 2TB drives are supported for PS4000 and PS6000 Series arrays.
Note:PS6500 Series arrays already supported these drives prior to this firmware release.
Performance has been improved for environments with large numbers of concurrent connections. The following new connection limits apply:
•The number of volumes that can be configured for replication has been increased from 128 to 256 for all group types except groups containing only PS4000 systems. For groups containing only PS4000 arrays, the limit remains at 32.
•Up to 1024 connections per pool are supported for all groups except groups containing only PS4000 Series arrays, providing increased scalability in virtualized environments. These connections may be distributed across a maximum of 512 volumes. Using more than 512 concurrent connections per pool requires increasing initiator timeout values to avoid initiator retries and possible loss of connectivity during failovers.
•For groups containing only PS4000 arrays, 256 connections per pool are supported.
•The supported number of SCSI-3 Persistent Reservation registrants per volume has been raised to 96 to support larger clusters.
Note:See Table 2, Supported Configuration Limits, for a complete listing of supported limits available here.
Group Manager GUI Improvements
The Group Manager GUI has several usability enhancements:
•The Group, Volumes, Replication, and Monitoring branches in the navigation tree can be quickly accessed with buttons at the bottom of the window.
•The display of status information for alarms and system operations has been improved.
•Context-sensitive online help has been added.
•Improvements have been made to the GUI to improve accessibility for visually-impaired users.
•The ability to pause and resume manual replication has been added to the GUI.
Event Logging
•A failed spare now generates a health condition indicating that it should be replaced.
•Opening the cover of a PS5500, PS6500 or PS6510 now generates an INFO level message in the event log instead of a WARNING message. If the cover remains open for more than 5 minutes, a CRITICAL message is issued.
•If an administrator attempts to delete a member that is offline but not vacated, a warning message is displayed requesting confirmation of the action.
Support for vStorage APIs for Array Integration
Beginning with version 5.0, the PS Series Array Firmware supports VMware vStorage APIs for Array Integration (VAAI) for VMware vSphere 4.1 and later. The following new ESX functions are supported:
•Hardware Assisted Locking - Provides an alternative means of protecting VMFS cluster file system metadata, improving the scalability of large ESX environments sharing datastores.
•Block Zeroing - Enables storage arrays to zero out a large number of blocks, speeding provisioning of virtual machines.
•Full Copy - Enables storage arrays to make full copies of data without requiring the ESX Server to read and write the data.
VAAI provides hardware acceleration for datastores and virtual machines residing on array storage, improving performance with the following:
•Creating snapshots, backups, and clones of virtual machines
•Using Storage vMotion to move virtual machines from one datastore to another without storage I/O
•Data throughput for applications residing on virtual machines using array storage
•Simultaneously powering on many virtual machines
•Refer to the VMware documentation for more information about vStorage and VAAI features.
Formatting of E-Mail Event Notifications
When sending email to an administrator, the Group previously sent a message with the email subject in the format of: n events from group group_name, which is difficult to sort. The format of the email subject has been changed to: group_name: n events for better clarity and ease of sorting.
Telnet, SSH, and FTP Security Enhancements
The following security enhancements have been made surrounding telnet, SSH, and FTP access to the Group Manager GUI and CLI:
•Telnet, SSH Version 1.0 and FTP services are now fully shut down when disabled by an administrator using the CLI.
•A new Python module has been added to the Host Scripting Toolkit (HST) to support automated interaction with the EqualLogic CLI via scripts using SSH. This Python-based toolkit addition also supports telnet, and the existing Perl based tools continue to be supported for customers using the Perl HST modules.
RAID Reconstruction Performance Improvements
•Enhanced prioritization algorithms that improve RAID reconstruction performance under certain heavy-load conditions, previously applied to 1TB drives, are now applied to all drives.
For complete details see the PS Series Storage Arrays: Release Notes Version 5.0.2