ServeRAID M5025 SAS/SATA Controller for IBM System x

The ServeRAID M5025 SAS/SATA Controller for IBM System x is the next-generation 6 Gbps SAS 2.0 external RAID controller. This adapter has two external mini-SAS connectors to control up to 240 devices and supports global and dedicated hot-spare drives. The ServeRAID M5025 controller delivers a full 6-Gbps SAS 2.0 RAID solution with the performance advantages of an extended cache and a standard battery backup unit. With the attachment of the ServeRAID M5000 Advanced Feature Key, the ServeRAID M5025 offers the option of RAID 6, RAID 60 and SED Encryption Key Management.

Specifications

The ServeRAID M5025 adapter card has the following specifications:

• Eight external 6 Gbps SAS 2.0 ports implemented via two four-lane (x4) connectors
• Two mini-SAS external connectors (SFF-8088)
• 6 Gbps throughput per SAS port
• 800 MHz PowerPC processor with LSI SAS2108 6 Gbps RAID on Chip (ROC) controller
• PCI Express 2.0 x8 host interface
• 512 MB onboard data cache (DDR2 running at 800 MHz)
• Intelligent lithium polymer battery backup unit standard with up to 48 hours of data retention
• Supports RAID levels 0, 1, 5, 10, and 50 (RAID 6 and 60 support with the optional M5000 Advanced Feature Key)
• Connects to
  • Up to 240 SAS or SATA drives
  • Up to 9 daisy-chained enclosures per port
• SAS and SATA drives are supported, but the mixing of SAS and SATA in the same RAID array is not supported
• Supports up to 64 logical volumes
• Supports LUN sizes up to 64 TB
• Configurable stripe size up to 1 MB
• Compliant with Disk Data Format (DDF) configuration on disk (COD)
• S.M.A.R.T. support

The key difference between the ServeRAID M5025 and M5015 RAID controllers is that the M5025 has two external SAS 2.0 x4 connectors and the M5015 has two internal SAS 2.0 x4 connectors.

Features

The ServeRAID M5025 adapter card has the following features:

• Auto-resume on array rebuild or array reconstruction after loss of system power

  Auto-resume uses non-volatile NVRAM to save rebuild progress during a host reboot or power failure to automatically resume from the last checkpoint. Auto-resume will ensure data integrity is maintained through the process. The card supports a number of features that can be implemented without rebooting the server. Applications such as email and web server benefit from avoiding downtime during transition.

• Online Capacity Expansion

  Online Capacity Expansion (OCE) allows the capacity of a virtual disk to be expanded by adding new physical disks or making use of unused space on existing disks, without requiring a reboot.

• Online RAID Level Migration

  Online RAID Level Migration (also known as logical drive migration) provides the ability to migrate a virtual disk from any RAID level to any other RAID level without requiring a reboot. System availability and application functionality remain unaffected.

• Fast initialization for quick array setup
• Consistency check for background data integrity

  Consistency Check verifies that all stripes in a virtual disk with a redundant RAID level are consistent. For RAID 1, the consistency check will mirror data when an inconsistent stripe is detected. For RAID 5 or RAID 6, the consistency check will recreate the parity from the peer disks. Consistency checks can be scheduled to take place periodically.

• Extensive online configuration options: advanced monitoring and event notification

  Management tools provide convenience for configuration of logical volumes and alerts when errors have occurred or are about to occur.

• Patrol Read for media scanning and repairing

  Patrol read is a background sentry service designed to proactively discover and correct media defects (bad sectors) that arise normally as a disk drive ages. The service issues a series of verify commands and if a bad block is discovered, the card's firmware will use RAID algorithms to recreate the missing data and remap the sector to a good sector. The task is interruptible based on controller activity and host operations. The firmware also provides an interface where the patrol read task can be initiated, set up for continuous operation, and terminated from a management application. Patrol read can be activated by manual command or automatically.

• Global and dedicated Hot Spare with Revertible Hot Spare support:

  A hot spare rebuilds data from all virtual disks within the disk group in which it is configured. ServeRAID provides the ability to define a physical disk as a hot spare to replace a failed drive. Hot spares can be configured as either global or dedicated. A global hot spare allows any physical drive to be designated as a hot spare. A dedicated hot spare will allow the user to assign a hot spare drive to a particular array of the same drive type.