One SCSI LSI53C1020 controller for Ultra320 support
One SCSI LSI53C1030 controller for Ultra320 support
One SCSI LSI53C1030 controller for Ultra320 support
SCSI data transfer rate
Up to 320 MB/sec per channel
Up to 320 MB/sec per channel
Up to 320 MB/sec per channel
SCSI bus
LVD, Single-ended (SE)
LVD, Single-ended (SE)
LVD, Single-ended (SE)
SCSI termination
Active
Active
Active
Termination disable
Automatic through cable and device detection
Automatic through cable and device detection This is automatic capable, but jumpers by default do not allow auto termination on PERC 4/DC.
Automatic through cable and device detection
Devices per SCSI channel
Up to 15 Wide SCSI devices
Up to 15 Wide SCSI devices
Up to 15 Wide SCSI devices
SCSI device types
Synchronous or asynchronous
Synchronous or asynchronous
Synchronous or asynchronous
RAID levels supported
0, 1, 5, 10, 50
0, 1, 5, 10, 50
0, 1, 5, 10, 50
SCSI connectors
One 68-pin internal high-density connector for SCSI devices. One very high density 68-pin external connector for Ultra320 and Wide SCSI.
Two 68-pin internal high-density connectors for SCSI devices. Two very high density 68-pin external connectors for Ultra320 and Wide SCSI.
Two 68-pin internal high-density connectors for SCSI devices. Two very high density 68-pin external connectors for Ultra320 and Wide SCSI.
Serial port
3-pin RS232C-compatible connector (for manufacturing use only)
3-pin RS232C-compatible connector (for manufacturing use only)
3-pin RS232C-compatible connector (for manufacturing use only)
NOTE: PERC 4 controller cards are not PCI Hot Pluggable. The system must be powered down in order to
change or add cards.
Cache Memory
64 MB of cache memory resides in a memory bank for PERC 4/SC and 128 MB for PERC 4/DC and PERC 4e/DC. The RAID controller supports write-through or write-back caching, selectable for each logical drive. To improve performance in sequential disk accesses, the RAID controller uses read-ahead caching by default. You can disable read-ahead caching.
Onboard Speaker
The RAID controller has a speaker that generates audible warnings when system errors occur. No management software needs to be loaded for the speaker to work.
Alarm Beep Codes
The purpose of the alarm is to indicate changes that require attention. The following conditions trigger the alarm to sound:
A logical drive is offline
A logical drive is running in degraded mode
An automatic rebuild has been completed
The temperature is above or below the acceptable range
The firmware gets a command to test the speaker from an application
Each of the conditions has a different beep code, as shown in Table 2-2. Every second the beep switches on or off per the pattern in the code. For example, if the logical drive goes offline, the beep code is a three second beep followed by one second of silence.
Table 2-2. Alarm Beep Codes
Alarm Description
Code
A logical drive is offline.
Three seconds on, one second off
A logical drive is running in degraded mode.
One second on, one second off
An automatic rebuild has been completed.
One second on, three seconds off
The temperature is above or below the acceptable range.
Two seconds on, two seconds off
The firmware gets a command to test the speaker from an application.
Four seconds on
BIOS
For easy upgrade, the BIOS resides on 1 MB flash memory. It provides an extensive setup utility that you can access by pressing <Ctrl><M> at BIOS initialization to run the BIOS Configuration Utility.
Background Initialization
Background initialization is the automatic check for media errors on physical drives It ensures that striped data segments are the same on all physical drives in an array.
The background initialization rate is controlled by the rebuild rate set using the BIOS Configuration Utility, <Ctrl><M>. The default and recommended rate is 30%. Before you change the rebuild rate, you must stop the background initialization or the rate change will not affect the background initialization rate. After you stop background initialization and change the rebuild rate, the rate change takes effect when you restart background initialization.
NOTE: Unlike initialization of logical drives, background initialization does not clear data from the drives.
Configuration Features
Table 2-3 lists the configuration features for the RAID controller.
Table 2-3. Configuration Features
Specifications
PERC 4/SC
PERC 4/DC
PERC 4e/DC
RAID levels
0, 1, 5, 10, and 50
0, 1, 5, 10, and 50
0, 1, 5, 10, and 50
SCSI channels
1
2
2
Maximum number of drives per channel
14
14 (for a maximum of 28 on two channels)
14 (for a maximum of 28 on two channels)
Array interface to host
PCI Rev 2.2
PCI Rev 2.2
PCI Express Rev. 1.0a
Cache memory size
64 MB SDRAM
Up to 128 MB SDRAM
Up to 128 MB SDRAM
Cache Function
Write-back, write-through, adaptive read-ahead, non read-ahead, read-ahead
Write-back, write-through, adaptive read-ahead, non read-ahead, read-ahead
Write-back, write-through, adaptive read-ahead, non read-ahead, read-ahead
Number of logical drives and arrays supported
Up to 40 logical drives and 32 arrays per controller
Up to 40 logical drives and 32 arrays per controller
Up to 40 logical drives and 32 arrays per controller
Only SCSI accessed fault-tolerant enclosure (SAF-TE) and SES
Only SAF-TE and SES
Only SAF-TE and SES
Mixed capacity hard drives
Yes
Yes
Yes
Number of 16-bit internal connectors
1
2
2
Cluster support
No
Yes
Yes
1 Hot swap of drives must be supported by enclosure or backplane.
Firmware Upgrade
You can download the latest firmware from the Dell website and flash it to the firmware on the board. Perform the following steps to upgrade the firmware:
Download the latest firmware and driver to a diskette.
The firmware is an executable file that downloads the files to the diskette in your system.
Restart the system and boot from the diskette.
Run pflash to flash the firmware.
NOTICE: Do not flash the firmware while performing a background initialization or data consistency
check, as it can cause the procedures to fail.
SMART Hard Drive Technology
The Self-Monitoring Analysis and Reporting Technology (SMART) detects predictable hard drive failures. SMART monitors the internal performance of all motors, heads, and hard drive electronics.
Drive Roaming
Drive roaming occurs when the hard drives are changed to different channels on the same controller. When the drives are placed on different channels, the controller detects the RAID configuration from the configuration information on the drives.
Configuration data is saved in both non-volatile random access memory (NVRAM) on the RAID controller and on the hard drives attached to the controller. This maintains the integrity of the data on each drive, even if the drives have changed their target ID. Drive roaming is supported across channels on the same controller, except when cluster mode is enabled.
NOTE: Drive roaming does not work if you move the drives to a new controller and put them on different
channels. If you put drives on a new controller, the controller must have a clear configuration. In addition,
the drives must be on the same channel/target as they were on the previous controller to keep the same
configuration.
NOTE: Before performing drive roaming, make sure that you have first powered off both your platform
and your drive enclosure.
Table 2-4 lists the drive roaming features for the RAID controller.
Table 2-4. Features for Drive Roaming
Specification
PERC 4/SC
PERC 4/DC
PERC 4e/DC
Online RAID level migration
Yes
Yes
Yes
RAID remapping
Yes
Yes
Yes
No reboot necessary after capacity extension
Yes
Yes
Yes
Drive Migration
Drive migration is the transfer of a set of hard drives in an existing configuration from one controller to another. The drives must remain on the same channel and be reinstalled in the same order as in the original configuration.
NOTE: Drive roaming and drive migration cannot be supported at the same time. PERC can support
either drive roaming or drive migration at any one time, but not both at the same time.
Hardware Architecture Features
Table 2-5 displays the hardware architecture features for the RAID controller.
Table 2-5. Hardware Architecture Features
Specification
PERC 4/SC
PERC 4/DC
PERC 4e/DC
Processor
Intel GC80302 (Zion Lite)
Intel GC80303 (Zion)
80332
SCSI controller(s)
One LSI53C1020 Single SCSI controller
One LSI53C1030 Dual SCSI controller
One LSI53C1030 Dual SCSI controller
Size of flash memory
1 MB
1 MB
1 MB
Amount of NVRAM
32 KB
32 KB
32 KB
Hardware exclusive OR (XOR) assistance
Yes
Yes
Yes
Direct I/O
Yes
Yes
Yes
SCSI bus termination
Active or LVD
Active or LVD
Active or LVD
Double-sided dual inline memory modules (DIMMs)
Yes
Yes
Yes
Support for hard drives with capacities of more than eight gigabytes (GB)
Yes
Yes
Yes
Hardware clustering support on the controller
No
Yes
Yes
LED Operation
After you remove a physical drive and place it back in the slot for a rebuild, the LED blinks for the drive as it is being rebuilt.
Array Performance Features
Table 2-6 displays the array performance features for the RAID controller.
Table 2-6. Array Performance Features
Specification
PERC 4/SC, PERC 4/DC, and PERC 4e/DC
PCI host data transfer rate
2 - 4 GB/sec, depending on the system
Drive data transfer rate
Up to 320 MB/sec
Maximum size of I/O requests
6.4 MB in 64 KB stripes
Maximum queue tags per drive
As many as the drive can accept
Stripe sizes
8 KB, 16 KB, 32 KB, 64 KB, or 128 KB
Maximum number of concurrent commands
255
Support for multiple initiators
Only on PERC 4/DC and PERC 4e/DC
Fault Tolerance Features
Table 2-7 describes the fault tolerance capabilities of the RAID controller.
1 The length of data retention depends on the cache memory configuration.
Operating System Software Drivers
Operating System Drivers
Drivers are provided to support the controller on the following operating systems:
Windows® 2000
Windows 2003
Novell® NetWare®
Red Hat® Linux, Advanced Server, Enterprise
NOTE: We support both 32-bit (x86) and 64-bit (IA64) processors for Windows 2003 and Red Hat Linux.
See the CERC and PERC RAID Controllers Operating System Driver Installation Guide for more information about the drivers.
SCSI Firmware
The RAID controller firmware handles all RAID and SCSI command processing and supports the features described in Table 2-8.
Table 2-8. SCSI Firmware Support
Feature
PERC 4/SC, PERC 4/DC, and PERC 4e/DC Description
Disconnect/reconnect
Optimizes SCSI bus utilization
Tagged command queuing
Multiple tags to improve random access
Multi-threading
Up to 255 simultaneous commands with elevator sorting and concatenation of requests per SCSI channel
Stripe size
Variable for all logical drives: 8 KB, 16 KB, 32 KB, 64 KB, or 128 KB.
Rebuild
Multiple rebuilds and consistency checks with user-definable priority.
RAID Management Utilities
Software utilities enable you to manage and configure the RAID system, create and manage multiple disk arrays, control and monitor multiple RAID servers, provide error statistics logging, and provide online maintenance. The utilities include:
BIOS Configuration Utility
Dell Manager for Linux
Dell OpenManage™ Array Manager for Windows and Netware
BIOS Configuration Utility
The BIOS Configuration Utility configures and maintains RAID arrays, clears hard drives, and manages the RAID system. It is independent of any operating system. See BIOS Configuration Utility and Dell Manager for additional information.
Dell OpenManage Array Manager is used to configure and manage a storage system that is connected to a server, while the server is active and continues to handle requests. Array Manager runs under Novell NetWare, Windows 2000, and Windows Server 2003. Refer to the online documentation that accompanies Array Manager or the documentation section at
support.dell.com for more information.
NOTE: You can run the OpenManage Array Manager remotely to access NetWare, but not locally.
Server Administrator Storage Management Service
Storage Management provides enhanced features for configuring a system's locally attached RAID and non-RAID disk storage. Storage Management runs under Red Hat Linux, Windows 2000, and Windows Server 2003. Refer to the online documentation that accompanies Storage Management or the documentation section at
support.dell.com for more information.
