Acer SCSI 320-0 User Manual

HARDWARE

GUIDE

MegaRAID® SCSI 320-0 Zero-Channel RAID Controller

September 2002

®

This document contains proprietary information of LSI Logic Corporation. The information contained herein is not to be used by or disclosed to third parties without the express written permission of an officer of LSI Logic Corporation.

LSI Logic products are not intended for use in life-support appliances, devices, or systems. Use of any LSI Logic product in such applications without written consent of the appropriate LSI Logic officer is prohibited.

Document DB15-000258-00, First Edition (September 2002) This document describes the initial release of LSI Logic Corporation’s MegaRAID SCSI 320-0 Controller and will remain the official reference source for all revisions/releases of this product until rescinded by an update.

LSI Logic Corporation reserves the right to make changes to any products herein at any time without notice. LSI Logic does not assume any responsibility or liability arising o ut of the app lication o r use of any p roduct des cribed her ein, except as expressly agreed to in writing by LSI Logic; nor does the purchase or use of a product from LSI Logic convey a license under any patent rights, copyrights, trademark rights, or any other of the intellectual property rights of LSI Logic or third parties.

LSI Logic, the LSI Logic logo design, and MegaRAID are trademarks or registered trademarks of LSI Logic Corporation. Intel is a registered trademark of Intel Corporation. MS-DOS, Microsoft Windows and Windows NT are registered trademarks of Microsoft Corporation. UNIX is a registered trademark of the Open Group. Novell NetWare is a registered trademark of Novell Corporation. ASPI is a registered trademark of Adaptec, Inc. All other brand and product names may be trademarks of their respective companies.

CD

To receive product literature, visit us at http://www.lsilogic.com. For a current list of our distributors, sales offices, and design resource

centers, view our web page located at http://www.lsilogic.com/contacts/na_salesoffices.html

ii

FCC Regulatory Statement

This device complies with Part 15 of the FCC Rules. Operation is subject to the following two conditions: (1) this device may not cause harmful interference, and (2) this device must accept any interference received, including interference that may cause undesired operation.

Warning: Changes or m odificati ons to this un it not expressly approved by

This equipment has been tested and found to comply with the limits for a Class B digital device, pursuant to Part 15 of the FCC Rules. These limits are designed to provide reasonable protection against harmful interference in a residential installation. This equipment generates, uses and can radiate radio frequency energy and, if not installed and used in accordance with the instructions, may cause harmful interference to radio communications. However, there is no guarantee that interference will not occur in a specific installation. If this equipment does cause harmful interfere nc e to r adi o o r tel evision receptio n, wh ic h can be determined by turning the equipment off and on, try to correct the interference by one or more of the following measures:

the party responsible for compliance could void the user's authority to operate the equipment.

1. Reorient or relocate the receiving antenna.

2. Increase the separation between the equipment and the receiver.

3. Connect the equipment into an outlet on a circuit different from that to which the receiver is connected.

4. Consult the dealer or an experienced radio/TV technician for help.

Shielded interface cables must be used with this product to ensure compliance with the Class B FCC limits.

Model Number: Series 520

Disclaimer – LSI LOGIC certifies on ly that this product wi ll w ork c orrec tl y when

this product is used with the same jumper settings, the same system configuration, the same memory module parts, and the same peripherals that were tested by LSI LOGIC with this product. The complete list of tested jumper settings, system configurations, peripheral devices, and memory modules are documented in the LSI LOGIC Compatibility Report for this product. Call your LSI LOGIC sales r epresentative for a copy of the Compati bility Repor t for this product.

iii

iv

Preface

Audience

Organization

This book is the primary reference and Hardware Guide for the LSI Logic MegaRAID the MegaRAID controller and for configuring RAID arrays. It also contains background information on RAID.

The MegaRAID SCSI 3 20-0 controller suppor ts single-en ded and lowvoltage differential (LVD) SCSI devices on an Ultra320 and Wide SCSI channel with data tran sfer rates up to 320 Mbytes/s.

This document is intende d for people who need to insta ll the MegaRA ID SCSI 320-0 Controlle r in a server and then cre ate and configure RAID arrays.

This document ha s the following chapters and appen dixes:

®

SCSI 320-0 Controller. It contains instr uctions for installing

· Chapter 1, Overview, provides an overview of the MegaRAID S CSI

320-0 and basic S CSI features.

· Chapter 2, Introduction to RAID, introduces impor tant RAID

concepts.

· Chapter 3, RAID Levels, describes each sup ported RAID l evel and

the factors to consider when choosing a RAID level.

· Chapter 4, Features, explains the features of the M egaRAI D® SCS I

320-0.

· Chapter 5, Configuring Physical Drives, Arrays, and Logical

Drives, explains how to configure SCSI physical drives, arrays, and

logical drives.

MegaRAID SCSI 320-0 Hardware Guide v

Technical Support

· Chapter 6, Hardware Installation, explains how to install the

MegaRAID SCSI 320-0 co ntroller.

· Chapter 7, Troubleshooting, provides troubleshooting informatio n

for the MegaRAID SCSI 320-0 c ontroller.

· Appendix A, Audible Warnings, explains the meaning of the

warning tones generated by the MegaRAID SCSI 320-0 RAID controller.

If you need help installing, c onfiguring, or r unning the MegaRAID SCSI 320-0 Controller, you may be able to find the information you need at the MegaRAID suppor t page at http://megaraid.lsilogic.com

If this does not resolve your problem, you can call your LSI Logic OEM Tec hnical Suppor t representative at 678-728- 1250. Before you call, please complete the M egaRAID Problem Report for m.

MegaRAID Problem Report Form

Customer Information MegaRAID Information

Name: Today’s Date: Company: Date of Purchase: Address: Invoice Number: City/State: Serial Number: Country: Email Address: Cache Memory: Phone: Firmware Version: Fax : BI OS Ver s io n:

System Information

Motherboard: BIOS manufacturer: Operating System: BIOS Date: Op. Sys. Ver.: Video Adapter:

vi Preface

MegaRAID Problem Report Form (Cont.)

MegaRAID Driver Ver.:

Network Card: System Memor y: Other disk controllers

installed:

Description of problem:

Steps necessary to re-create problem:

1.

2.

3.

4.

Logical Drive Configuration

Use this form to reco rd the configuration detail s for your logical drives.

Logical Drive Configuration

CPU Type/Speed:

Other adapter cards Installed:

Logical

Drive

LD0 LD1 LD2 LD3 LD4 LD5 LD6 LD7 LD8

RAID Level

Stripe

Size

Preface vii

Logical Drive

Size

Cache Policy

Read

Policy

Write

Policy

# of Physical

Drives

Logical Drive Configuration (Cont.)

Logical

Drive

LD9 LD10 LD11 LD12 LD13 LD14 LD15 LD16 LD17 LD18 LD19 LD20 LD21 LD22

RAID Level

Stripe

Size

Logical Drive

Size

Cache Policy

Read

Policy

Write

Policy

# of Physical

Drives

LD23 LD24 LD25 LD26 LD27 LD28 LD29 LD30 LD31 LD32 LD33

viii Preface

Logical Drive Configuration (Cont.)

Logical

Drive

LD34 LD35 LD36 LD37 LD38 LD39

RAID Level

Stripe

Size

Physical Device Layout

Use this form to reco rd the physical device layout.

Physical Device Layout

Target ID Device type

Logical Drive

Size

Cache Policy

Read

Policy

Channel 0 Channel 1

Write

Policy

# of Physical

Drives

Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number

Preface ix

Physical Device Layout (Cont.)

Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number

Channel 0 Channel 1

Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number

xPreface

Physical Device Layout (Cont.)

Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number

Channel 0 Channel 1

Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number

Preface xi

Physical Device Layout (Cont.)

Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number

Channel 0 Channel 1

Firmware level

xii Preface

Chapter 1 Overview

MegaRAID SCSI 320-0 can be install ed in a special P CI expansion slot in a computer wi th a m otherboar d tha t has the LS I Logi c 53C 1030 SCSI chip. The zero-channel ad apter convert s a mother board’s own on-board SCSI controller into a powerful PCI SCSI RAID sol ution, and is targete d for 1U and 2U server environme nts.

The MegaRAID SCSI 3 20-0 is part of th e LSI Logic Intel i960RM/RS based MegaRAID contro ller family, and is an entr y-level to mid-range RAID controller soluti on. This controller offers a cos t-effective way to implement RAID in a ser ver.

The MegaRAID SCSI 320- 0 supports d ata transfer rates up to 320 Mbytes/s per channel, and it suppor ts up to fifteen non-U ltra SCSI devices. MegaRAID SCSI 320-0 includes MegaRAID features and performance.

MegaRAID SCSI 320-0 Hardware Guide 1-1

1.1 Features

MegaRAID SCSI 320-0 features inc lude:

· Provides a high performan ce I/O migration path while pr eserving

existing PCI-SCSI software.

· Performs SCSI data transfers up to 320 Mbytes/s.

· Performs synchronous operation on a wi de low-voltage differential

(LVD) SCSI bus.

· Allows up to 15 LVD SCSI devices on the wi de bus.

· Includes an Intel GC8 0302 integrated I/O processor that performs

RAID calculations an d routing.

· Supports 32 to 128 Mbytes of SDRAM on-board cache memory used

for read and write-back cachi ng, and RAID 5 parity g eneration.

1.2 NVRAM and Flash ROM

A 32 Kbyte x 8 nonvolatile random access memor y (NVRAM) stores RAID system configura tion information. The Mega RAID SCSI 320-0 firmware is stored in flash ROM for easy upgrade.

1.3 Single-Ended and Differential SCSI Buses

The SCSI standard de fines two electrical buses:

· Single-ended bus

· Low-voltage differential bus

1-2 Overview

1.4 Maximum Cable Length for SCSI Standards

Table 1.1 lists the maximum SCSI cable length and number of disk drives

that you can use, depending o n the SCSI speeds.

Table 1.1 Maximum Cable Length for SCSI Standards

Single Ended

Standard

Ultra SCSI 1.5 m 12 m 7 Ultra SCSI 3 m 12 m 3 Wide Ultra SCSI 12 m 15 Wide Ultra S CSI 1.5 m 12 m 7 Wide Ultra SCSI 3 m 12 m 3 Ultra 2 SCSI 25 m 1 Ultra 2 SCSI 12 m 7 Wide Ultra 2 SCSI 25 m 1 Wide Ultra 2 SCSI 12 m 15 Ultra160 SCSI 25 m 1 Ultra160 SCSI 12 m 7 Wide Ultra160 SCSI 25 m 1 Wide Ultra160 SCSI 12 m 15 Ultra320 12 m 15

SCSI

Low-Voltage

Differential SCSI

Maximum # of

Drives

Ultra320 20 m 1

Maximum Cable Length for SCSI Standards 1-3

1.5 SCSI Bu s Widths and Maximum Throughput

Table 1.2 lists the SCSI bus widths and maximum throug hput, based on

the SCSI speeds.

Table 1.2 SCSI Bus Widths and Maximum Throughput

SCSI Standard SCSI Bus Width SCSI Throughput

Fast Wide SCSI 16 bits 20 Mbytes/s Wide Ultra S CSI 16 bits 40 Mbytes/s Wide Ultra 2 SCSI 16 bits 80 Mbytes/s Wide Ultra 160 SCSI 16 bits 160 Mbytes/s Ultra 320 SCSI 16 bits 320 Mbytes/s

1.6 Documentation

The MegaRAID SCSI 3 20-0 documentation se t includes:

· The MegaRAID SCS I 320-0 ZCR Hardware Guide

· The MegaRAID Con figuration Software Guide

· The MegaRAID Operating System Driver Instal lation Guide

1.6.1 MegaRAID SCSI 320-0 ZCR Hardware Guide

The Hardware Guide contai ns the RAID overview, RAID planning, and RAID system configura tion information you need firs t. Read this document first.

1.6.2 MegaRAID Configuration Software Guide

This manual descr ibes the software configura tion utilities that you can use to configure and m odify RAID systems.

1-4 Overview

1.6.3 MegaRAID Operating System Driver Installation Guide

This manual provides detailed information about installing the MegaRAID SCSI 320-0 operating s ystem drivers.

Documentation 1-5

1-6 Overview

Chapter 2 Introduction to RAID

This chapter introdu ces importan t RAID concepts. It contains th e following sections:

· Section 2.1, “R AID Benefits”

· Section 2.2, “MegaRAI D SCSI 320-0 – Host-Ba sed RAID Solution”

· Section 2.3, “RAID Overview”

RAID (Redundant Array of Independent Di sks) is a dat a storage meth od in which data, along with par ity information, is distr ibuted among two or more hard disks (called an array) to improve performance and reliability. The RAID array appears to the host computer as a single storage unit or as multiple logical u nits. I/O is expedited because several disks can be accessed simultan eously. RAID system s provide improved data s torage reliability and fault tolerance compared to single-dr ive computers. If a disk drive in a RAID a rray fails, data can be recons tructe d from th e data and parity inform ation on the remaining disk drives.

2.1 RAID Benefits

RAID is widely used be ca us e it im pr oves I/O performanc e a nd i nc reas es storage subsystem r eliability. RAID provides data se curity through fault tolerance and redundant dat a storage. The MegaRAID SCSI 320-0 management software co nfigures and monitors RAID d isk arrays.

2.1.1 Improved I/O

Although disk dr ive capabilities have improved drastically, actual performance of individ ual disk drives has been i mproved only three to four times in the last decade. RAI D provides a way to achieve much better data throughput.

MegaRAID SCSI 320-0 Hardware Guide 2-1

2.1.2 Increased Reliability

The electromechan ical components of a d isk subsystem operate mor e slowly, require more power, and generate more noise and vibra tion than electronic devices. These factors reduce the reliab ility of data stored on disks. RAID provides a way to achieve much better fault tolerance and data reliability.

2.2 MegaRAID SCSI 320-0 – Host-Based RAID Solution

RAID products are ei ther host-based or external. The MegaRAID SCSI 3 20-0 controller is a hos t-based RAID solutio n.

The MegaRAID SCSI 3 20-0 is a PCI adapter card that is installed in a PCI expansion slot in a host system with a motherboard that contains the LSI Logic 53C1030 S CSI chip.

2.2.1 Host-Based RAID

A host-based RAID pr oduct puts all of the RAID in telligence on an adapter card that is in stalled in a network se rver. A host-based RAID product provides the best performance. MegaRAID SCSI 320-0 is part of the file server, so it can transmit dat a directly across th e computer’s buses at data transfer speeds up to 1 32 Mbytes/s.

The available sequential data transfer rate is de termined by the following factors:

· The sustained data transfer rate on the mot herboard PCI bus

· The sustained data transfer rate on the i96 0RM PCI to PCI bridge

· The sustained data transfer rate of the SCSI controller

· The sustained data transfer rate of the SCSI devices

· The number of SCSI chan nels

· The number of SCSI disk d rives

Host-based solu tions must provide operating sy stem-specific dr ivers.

2-2 I ntroduction to RAID

2.2.2 SCSI-to-SCSI External RAID

A SCSI-to-SCSI extern al R AID produ ct puts the RA ID inte llige nce ins ide the RAID chassis and uses a plain SCSI hos t adapter installed in th e network ser ver. The data transfer rate is limited to the bandwidth of the SCSI channel. A SCSI-t o-SCSI extern al RAID product t hat h as two wid e SCSI channels operati ng at speeds up to 320 Mbytes /s must squeeze the data into a single wide SCSI (320 Mbytes/s) channel back to the host computer.

In SCSI-to-SCSI extern al RAID pr oducts, the di sk dr ive subsystem us es only a single SCSI ID, which allows you to connect multip le drive subsystems to a si ngle SCSI controller.

2.3 RAID Overview

RAID is a collection of specifications that describes a system for ensuring the reliability and stability of data stored on large disk subsy stems. A RAID system can be implemented in a number of different versions (or RAID levels). MegaRAID SCSI 320-0 supp orts standard RA ID levels 0, 1, and 5, and RAID levels 10 and 50, special RAID versions suppor ted by MegaRAID SCSI 320-0.

2.3.1 Physical Array

A RAID array is a collection of physical disk drives governed by the RAID management software. A RAID array appears to th e host computer as one or more logical drives.

2.3.2 Logical Drive

A logical drive is a par tition in a physical array of disks that is made up of contiguous data s egments on the physical disk s. A logical drive can consist of any of the following:

· An entire physical array

· More than one entire p hysical array

· A part of an array

· Parts of more than on e array

RAID Overvi ew 2-3

· A combination of any two of the above conditions

2.3.3 Consistency Check

A consistency check verifies the correctness of redundant data in a RAID array. For example, in a system with d istributed parity, checking consistency means c omputing the parit y of the data drives and comparing the results to the contents of the parity dri ves.

2.3.4 Fault Tolerance

Fault tolerance is achieved through auto detection of failed drives, rebuilds with hot spare drive, the ability to hot swap drives, Configuration on Disk capability, and support for self-monitor ing a nalys is and r eporting technology (SMART). MegaRAID SCSI 320- 0 provides hot swapping through the hot spare feature. A ho t spare drive is an unused onl ine available drive that MegaRAID SCSI 320-0 can i nstantly plug into the system when an active dr ive fails.

After the hot sp are is automa tically moved into the RAID subsy stem, the failed drive is automaticall y rebuilt. The RAID disk array continues to handle request while the rebuild occurs.

2.3.5 Disk Striping

Disk striping writes data across multiple disk drives instead of just one disk drive, as shown in Figure 2.1.

2-4 I ntroduction to RAID

Figure 2.1 Disk Striping

MegaRAID Controller

Disk striping involves partitioning each disk d rive’s storage space into stripes that can vary in size from 2 to 128 Kbytes. These stripes are interleaved in a repeated, sequ ential manner. The combined storage space is composed of s tripes from each dr ive. MegaRAID SCSI 320-0 suppor ts stripe sizes of 2, 4, 8, 1 6, 32, 64, or 128 Kbytes.

For example, in a four-disk system using on ly disk striping (as in RAID level 0), segment 1 is written to d isk 1, segment 2 is wri tten to disk 2, and so on. Disk stripi ng enhances p erformance be cause multip le dri ves are accessed simultane ously; but disk striping does not provide data redundancy.

2.3.5.1 Stripe Width

Stripe width i s a measure of the number of disks involved in an array where striping is implemented. For example, a four-disk array with disk striping has a st ripe width of four.

Segment 1 Segment 5 Segment 9

Segment 2 Segment 6

Segment 10

Segment 3 Segment 7

Segment 11

Segment 4 Segment 8

Segment 12

RAID Overvi ew 2-5

2.3.5.2 Stripe Size

The stripe s ize is the length of the interl eaved data segments that MegaRAID SCSI 320-0 wr ites across multipl e drives. MegaRAID SCSI 320-0 suppor ts stripe sizes o f 2, 4, 8, 16, 32, 64, or 128 K bytes.

2.3.6 Disk Mirroring

With disk mirroring (used in RAID 1), data wr itten to one disk dr ive is simultaneously wr itten to another disk dr ive, as shown in Figure 2.2.

Figure 2.2 Disk Mirroring

MegaRAID Controller

Segment 1 Segment 2 Segment 3 Segment 4 Segment 4 Duplicated

If one disk drive fails, the contents of the other disk dr ive can be used to run the system and reconstr uct the failed drive. The primary advantage of disk mirroring i s that it provides 100% data r edundancy. Since the contents of the disk d rive are completely wr itten to a second dri ve, it does not matter if o ne of the drives fails. Both drives contain th e same data at all times. Either drive can act as the operationa l drive.

Although disk mirror ing provides 100% redund ancy, it is expensive because each dr ive in the system must be duplic ated.

2-6 I ntroduction to RAID

Segment 1 Duplicated Segment 2 Duplicated Segment 3 Duplicated

2.3.7 Disk Spanning

Disk spanning allows multiple disk drives to function like one bi g drive. Spanning overcomes lack of disk s pace and simplifies storag e management by combinin g existing resources or adding rel atively inexpensive resources. For example, four 60 Gbyte disk drives can be combined to appear to t he operating system as one single 240 Gbyte drive.

Disk spanning alone does not provide reliability or performance enhancements. Spanned logical drives must have the same stripe s ize and must be contiguous. In Figure 2.3, two RAID 1 arrays are turned into a RAID 10 array.

Figure 2.3 Disk Spanning

MegaRAID Controller

Data Flow

Segment 1 Segment 3 Segment 5

RAID 1

Disk 1 Disk 4Disk 3

Segment 1 Segment 3 Segment 5

Disk 2

Segment 2 Segment 4 Segment 6

RAID 0

RAID 1

Segment 2

Segment 2 Segment 4

Segment 4 Segment 6

Segment 6

RAID Overvi ew 2-7

Table 2.1 describes how disk sp ann in g is us ed for RAID 10 an d RA ID 50 .

Table 2.1 Spanning for RAID 10 and RAID 50

Level Description

10 Configure RAID 10 by spanning two contiguous RAID 1 logical drives.

The RAID 1 logical drives must have the same stripe size.

50 Configure RAID 50 by spanning two contiguous RAID 5 logical drives.

The RAID 5 logical drives must have the same stripe size.

2.3.8 Parity

Parity generates a set of redunda ncy data f rom two or more par ent data sets. The redundancy data can be use d to reconstr uct one of the pa rent data sets. Parity data does not f ully duplicate the pa rent data sets. In RAID, this method is applied to entire dr ives (dedicated parity ) or to stripes acros s all disk drives in an array (distributed parit y).

RAID 5 combines distributed parity with disk striping. If a single disk drive fails, it can be rebuilt from the parity an d the data on the remaining drives. Parity provides redundancy for one dr ive failure without duplicating the c ontents of entire disk dr ives, but parity generation can slow the write process.

2.3.9 Hot Spares

A hot spare is an extra, unused disk drive that is par t of the disk subsystem. It is usually in standby mode, ready for service if a drive fails. Hot spares permit you to replace failed drives without syste m shutdown or user inter vention.

Note:

Spanning two conti guous RAID 0 logical d rives does not produce a new RAID level or add fault tolerance. It does increase the si ze of the logical volume and imp roves performance by doubling the number of spindles.

MegaRAID SCSI 320-0 implements automatic and transparent rebuilds using hot spare dr ives, providing a high degree of fault tolerance and zero downtime. The MegaRAID SCSI 320-0 RAID Management software allows you to specify physical drives as hot spares. When a hot spare is needed, the MegaRAID SC SI 320 -0 controller assigns the hot s pa re that

2-8 I ntroduction to RAID

has a capacity cl osest to and at least as great as that of the failed dr ive to take the place of the failed dr ive.

Note:

2.3.10 Hot Swapping

Hot swapping is the manual replaceme nt o f a defective physical disk u ni t while the computer is still runn ing. When a new dri ve has been instal led, you must issue a command to rebuild the drive.

2.3.11 Disk Rebuild

You rebuild a di sk drive by recreating the data tha t had been stored on the drive before the drive failed. Rebuilding can be done only in arrays with data redundanc y such as RAID level 1, 5, 10, and 50.

Standby (warm spare) r ebuild is employed in a mirrored (RAID 1) system. If a disk dr ive fails, an identical drive is imm ediately available. The primar y data source disk d rive is the original disk drive.

A hot spare can be us ed to rebuild disk drives in RAID 1, 5, 10 , or 50 systems. If a hot spare is not available, the failed disk drive must be replaced with a new disk drive so that the data on the failed drive can be rebuilt.

Hot spares are employed only in arrays with redundanc y— for example, RAID levels 1, 5, 10, and 50. A hot spare connected to a speci fic MegaRAID SCSI 320- 0 controller can be used only to r ebuild a d rive that i s co nnecte d to the same controller.

The MegaRAID SCSI 3 20-0 controller automatic ally and transparently rebuilds failed drives with user-def inable rebuild rates. If a hot spare is available, the rebuild starts automatically when a drive fails. MegaRAID SCSI 320-0 automati cally restar ts the system and the r ebuild if the system goes down during a rebuild.

2.3.11.1 Rebuild Rate

The rebuild rate is the fraction of the compute cycle s dedicated to rebuilding failed drives. A rebuild rate of 100% mea ns the system is totally dedicated to r ebuilding the failed drive.

RAID Overvi ew 2-9

The MegaRAID SCSI 32 0-0 rebuild rate can be configured between 0% and 100%. At 0%, the rebuild is only done if the system is not doing anything else. At 100%, the rebuild has a higher pr iority than any other system activity.

2.3.12 Logical Drive States

Table 2.2 describes the l ogical drive states.

Table 2.2 Logical Drive Sta tes

State Description

Optimal The drive operating condition is good. All configured drives are

Degraded The drive o perating cond ition is no t optimal. On e of the

Failed The drive has failed. Offline The drive is not available to MegaRAID SCSI 320-0.

2.3.13 SCSI Drive States

A SCSI disk dri ve can be in one of the states described in Table 2.3.

Table 2.3 SC SI Drive States

State Description

Online (ONLIN)

Ready (READY)

Hot Spare (HOTSP)

Fail (FAIL)

online.

configured drives has failed or is offline.

The drive is functioning normally and is a part of a configured logical drive.

The drive is functioning normally but is not part of a configured logical drive and is not designated as a hot spare.

The drive is powered up and ready for use as a spare in case an online drive fails.

A fault has occurred in the drive, placing it out of service.

Rebuild (REB)

2-10 Introduction to RAID

The drive is being rebuilt with data from a failed drive.

2.3.14 Disk Array Types

Table 2.4 describes the RAID dis k array types.

Table 2.4 Disk Array Types

Type Description

SoftwareBased

SCSI to SCSI The array controller resides outside of the host computer and

Bus-Based The array controller resides on the bus (for example, a PCI or

The array is managed by software running in a host computer using the host CPU bandwidth. The disadvantages associated with this method are the load on the host CPU and the need for different software for each operating system.

communicates with the host through a SCSI adapter in the host. The array management software runs in the controller. It is transparent to the host and independent of the host operating system. The disadvantage is the limited data transfer rate of the SCSI channel between the SCSI adapter and the array controller.

EISA bus) in the host comput er and has it s own CPU to generate the parity and handle other RAID functions. A busbased controller can transfer data at the speed of the host bus (PCI, ISA, EISA, VL-Bus) but is limited to the bus it is designed for. MegaRAID SCSI 320-0 resides on a PCI bus, which can handle data transfer at up to 132 Mbytes/s. With MegaRAID SCSI 320-0, the channel can handle data transfer rates up to 320 Mbytes/s per SCSI channel.

2.3.15 Enclosure Management

Enclosure managem ent is the intelligent m onitoring of the disk subsystem by software and/or hardware.

The disk subsystem can be part of the host computer or can be separate from it. Enclosure managem ent helps you stay informed of events in the disk subsystem, such as a drive or power supply failure. Enclosure management increa ses the fault tolerance of the disk subsystem.

RAID Overvi ew 2-11

2-12 Introduction to RAID

Chapter 3 RAID Levels

This chapter descr ibes each suppor ted RAID level and the factors to consider when choosi ng a RAID level. It contains the following sections:

· Section 3.1, “Selec ting a RAID Level”

· Section 3.2, “RAID 0”

· Section 3.3, “RAID 1”

· Section 3.4, “RAID 5”

· Section 3.5, “RAID 10”

· Section 3.6, “RAID 50”

3.1 Selecting a RAID Level

To en sure the best performanc e, you should select the optimal RAID level when you create a system drive. The optimal RA ID level for your disk array depends on a number of factors:

· The number of drives in the disk array

· The capacity of the dr ives in the array

· The need for data redundancy

· The disk performance requirements

Note:

MegaRAID SCSI 320-0 Hardware Guide 3-1

The SCSI 320-0 control ler suppor ts a maximum of 15 physical drives.

3.2 RAID 0

RAID 0 provides disk str iping across all dr ives in the RAID subsystem. RAID 0 does not provide any da ta redundancy, but does offer the best performance of any RAID level. RAID 0 breaks up data into sm aller blocks and then writes a block to each drive in the array . The size of each block is determined by the stri pe size parameter, set during the cr eation of the RAID set. RAID 0 of fers high bandwidth. By breaking up a l arge file into smaller blocks, MegaRAID SCS I 320-0 ca n use s everal drives to read or write the file faster. RAID 0 involves no parity cal culations to complicate the wri te operation. This makes RAID 0 ideal for application s that require high ban dwidth but do not require fault tolerance.

Uses RAID 0 provides high data throughput, especially for large

Strong Points Provides increased data throughput for large files. No

Weak Po in ts Does not provide fault tolerance. All data lost if any drive

Drives 1 to 30

files. Suitable for any environment that does not require fault tolerance.

capacity loss penalty for parity.

fails.

The MegaRAID SCSI 320-0 controller supports one or two SCSI channels, with up to 15 SCSI devices on each channel.

Figure 3.1 shows a RAID 0 array with four disk drives.

3-2 R AID Levels

Figure 3.1 RAID 0 Array

MegaRAID Controller

3.3 RAID 1

Segment 1 Segment 5 Segment 9

Segment 2 Segment 6

Segment 10

Segment 3 Segment 7

Segment 11

Segment 4 Segment 8

Segment 12

In RAID 1, the MegaRAID S CSI 320 -0 dup licates all d ata fro m one dr ive to a second drive. RAID 1 provides complete dat a redundancy, but at the cost of doubling the req uired data storage capaci ty.

Uses Use RAID 1 for small databases or any other environment

Strong Points Provides complete data redundancy. RAID 1 is ideal for any

Weak Po in ts R equires tw ice as many disk drives. Performance is impaired

Drives 2

that requi res fault tole rance but small capacity.

application that requires fault tolerance and minimal capacity.

during drive rebuilds.

Figure 3.2 shows a RAID 1 array.

RAID 1 3-3

Figure 3.2 RAID 1 Array

MegaRAID Controller

3.4 RAID 5

Segment 1 Segment 2 Segment 3 Segment 4 Segment 4 Duplicated

Segment 1 Duplicated Segment 2 Duplicated Segment 3 Duplicated

RAID 5 includes disk str iping at t he byte level and parity. In RAID 5, the parity information i s written to several drives. RAID 5 is best su ited for networks that perfor m many small I/O transac tions simultaneously.

RAID 5 addresses the bottlen eck issue for random I/O operation s. Since each drive contains both data and parity, numerous writes can take place concurrently. In addition, robust caching algorithms a nd hardware ba sed exclusive-or assist make RAID 5 perform ance exceptional in many different environments.

3-4 R AID Levels

Uses Provides high data throughput, especially for large files. Use

Strong Points Provides data redundancy and good performance in most

Weak Po in ts Disk drive performance is reduced if a drive is being rebuilt.

Drives 3 to 30

Figure 3.3 shows a RAID 5 array with six disk drives.

Figure 3.3 RAID 5 Array

RAID 5 for transaction process ing ap pl ica t io ns, because each drive can read and write independently. If a drive fails, the MegaRAID SCSI 320-0 uses the parity data to recreate all missing information. Use also for office aut oma tio n a nd online customer service that requires fault tolerance. Use for any application that has high read request rates but low write request rates.

environments

Environments with few processes do not perform as well because the RAID overhead is not offset by the performance gains in handling simultaneous processes.

The MegaRAID SCSI 320-0 controller supports one or two SCSI channels, with up to 15 SCSI devices on each channel.

Note: Parity is distributed

across drives in the array.

Disk 1 Disk 2 Disk 3 Disk 4 Disk 5 Dis k 6

Segment 1 Segment 7

Parity (9–12)

Segment 2 Segment 8

MegaRAID Controller

Data Flow

Segment 3 Segment 9

Parity (5–8)

Segment 4

Segment 10

Segment 5

Segment 11

Parity (1–4)

Segment 6

Segment 12

RAID 5 3-5

3.5 RAID 10

RAID 10 is a combin ation of RA ID 0 and RAID 1. RAI D 10 ha s mirrored drives. RAID 10 breaks up da ta into sm aller blocks, and then s tripes the blocks of data to each RAID 1 RAID set. Eac h RAID 1 RAID set then duplicates its data to its other drive. The size of each block is determined by the stripe size parameter, which is set during the creation of th e RA ID set. RAID 10 can sustain one to four drive failures while maintaining data integrity, if each failed disk is i n a different RAID 1 array.

Uses Works best for data storage that must have 100%

Strong Points Provides both high data transfer rates and complete data

Weak Po in ts Requires twice as many drives as all other RAID levels

Drives 4 to 18

redundancy of mirrored arrays and that also needs the enhanced I/O performance of RAID 0 (striped arr ays). RAID 10 works well for medium-sized databases or any environment that requires a higher degree of fault tolerance and moderate to medium capacity.

redundancy.

except RAID 1.

Must be a multiple of two.

Figure 3.4 shows a RAID 10 array with four disk dr ives.

3-6 R AID Levels

Figure 3.4 RAID 1 0 Array

MegaRAID Controller

Data Flow

3.6 RAID 50

Segment 1 Segment 3 Segment 5

RAID 1

Disk 1 Disk 4Disk 3

Segment 1 Segment 3 Segment 5

Disk 2

Segment 2 Segment 4 Segment 6

RAID 0

RAID 1

Segment 2

Segment 2 Segment 4

Segment 4 Segment 6

Segment 6

RAID 50 provides the features of both RAID 0 and RAID 5, including both parity and disk striping across mul tiple drives. RAID 50 is bes t implemented on two RAID 5 disk arrays with data striped across both disk arrays. RAID 50 breaks up data into smaller blocks, and then stripes the blocks of data to each RAID 5 RAI D set. RAID 5 bre aks up data into smaller blocks, calculates par ity by performing an exclusive-or on the blocks, and then writes the blocks of data and pa rity to each dr ive in the array. The s ize of each block is deter mined by the str ipe size parameter, which is set during the creation of the RAID set.

RAID 50 3-7

RAID 50 can sustai n one to four drive failures while maintaini ng data integrity, if each failed disk is i n a different RAID 5 array.

Uses Works best when used with data that requires high reliability,

Strong Points Provides high data throughput, data redundancy, and very

Weak Po in ts Requires 2 to 4 times as many parity drives as RAID 5. Drives 6 to 30

Figure 3.5 shows a RAID 50 array with si x disk drives.

Figure 3.5 RAID 5 0 Array

high request rates, high data transfer, and medium to large capacity.

good performance.

The MegaRAID SCSI 320-0 controller supports one or two SCSI channels, with up to 15 SCSI devices on each channel.

MegaRAID Controller

RAID 5

Disk 1 Disk 2 Disk 3

Segment 1 Segment 6

Parity (9-10)

Segment 2 Parity (5-6) Segment 9

3-8 R AID Levels

Parity (1-2) Segment 5

Segment 10

Data Flow

RAID 0

Disk 4 Disk 5 Disk 6

Segment 3 Segment 8

Parity (11-12)

RAID 5

Segment 4 Parity (7-8)

Segment 11

Parity (3-4) Segment 7

Segment 12

Chapter 4 Features

This chapter explains the features of the Me gaRAID SCSI 320-0. It contains the following sections :

· Section 4.1, “SMART Technology”

· Section 4.2, “Confi guration on Disk”

· Section 4.3, “Confi guration Features”

· Section 4.4, “Array Performance Features”

· Section 4.5, “RAID Management Features”

· Section 4.6, “Fault Tolerance Features”

· Section 4.7, “Soft ware Utilities”

· Section 4.8, “Operating System Software Drivers”

· Section 4.9, “Mega RAID SCSI 320-0 Speci fications”

· Section 4.10, “RAI D Management”

· Section 4.11, “Com patibility”

MegaRAID is a family of high perform ance intelligent PCI-to-SCSI h ost adapters with RAID control capabilities. MegaRAID SCSI 320-0 supports Ultra320 and Wide SCSI at data transfer rates up to 320 Mbytes/s, and up to 15 Wide devices and up to seven non-Wide devices.

4.1 SMART Technology

The MegaRAID SCSI 3 20-0 Self Monitorin g Analysis and Repor ting Tec hnology (SMART) detects up to 70% of all predictable dr ive failures. SMART monitors the internal performance of all motors, heads, and drive electronics.

MegaRAID SCSI 320-0 Hardware Guide 4-1

4.2 Configur ation on Disk

Configuration on Disk (d rive roaming) saves configuration informatio n both in nonvolatile random access memor y ( NVRAM) o n the Mega RAID SCSI 320-0, and on the di sk drives controlled by the MegaR AID SCSI 320-0. If the MegaRAID SCSI 320-0 is replaced, the new Mega RAID SCSI 320-0 controller can detect the actual RAID configuration, maintaining the integrity of the data on each drive, even if the drives have changed channel an d/or target ID.

4.3 Configur ation Features

Table 4.1 contains the configuration features for the MegaRAID 320-0.

Table 4.1 Configuration Features

Specification Feature

RAID levels 0, 1, 5, 10, and 50 SCSI channels 0 (The MegaRAID 320-0 enables

RAID capability on one or two SCSI channels controlled by the LSI Logic 53C1030 SCSI chip installed on the

motherboard.) Maximum number of drives per channel 15 Array interface to host PCI 2.2 Drive interface Fast and Wide, Ultra320 single-

Upgradeable cache size Cache memory onboard Cache function Write-through, write-back, adaptive

Multiple logical drives/arrays per controller

Maximum number o f MegaRAID SCSI 320-0 controllers per system

ended and low-voltage differential

(LVD)

read-ahead, no read-ahead, read-

ahead

Up to 40 logical drives per controller

1

4-2 Features

Table 4.1 Configuration Features (Cont.)

Specification Feature

Online capacity expansion Yes Hot spare support Yes Flashable fi rmware Yes Hot swap devices supported Yes Non-disk devices sup porte d Yes Mixed capacity hard disk drives Yes Suppor t for hard di sk drives with

capacities of more than 8 Gbytes. Clustering support (Failover control) No Online RAID level migration Yes RAID remapp ing Yes No reboot necessary after expansion Yes More than 20 0 Qtags per physic al drive Yes Hardware clustering support on the

board User-speci fied rebuild rate Yes

4.4 Array Performance Features

Table 4.2 lists the array performance features.

Table 4.2 Array Performance Features

Specification Feature

Yes

Host data transfer rate 533 Mbytes/s Drive data transfer rate 320 Mbytes/s Stripe sizes 2, 4, 8, 16, 32, 64, or 128 Kbytes

Array Performance Features 4-3

4.5 RAID Management Features

Table 4.3 lists the RAID management features.

Table 4.3 RAID Management Features

Specification Feature

Support for SNMP Yes Performance Monito r provided Yes Remote control and monitoring Yes Support for concurrent multiple stripe sizes Yes Windows NT, 2000, XP, and .NET server support using a GUI

client utility

4.6 Fault Tolerance Features

Table 4.4 lists the fault tolerance features.

Table 4.4 Fault Tolerance Features

Specification Feature

Support for SMART Yes Enclosure management SCSI-accessed fault-tolerant

Drive failure detection Automatic Manual disk replacement without system

shutdown Drive rebuild using hot spares Automatic Parity generation for RAID Hardware

Yes

enclosure (SAF-TE) compliant

Yes, by using hot swap features

4-4 Features

4.7 Software Utilities

Table 4.5 lists the software utilit y features.

Table 4.5 Software Utilit ies

Specification Feature

Graphical user interface Yes Management utility Yes Bootup configuration using MegaRAID Manager Yes Online read, write, and cache policy switching Yes

4.8 Operating System Software Drivers

MegaRAID SCSI 320-1 includes a DOS software configuration utility, and drivers for:

· Windows NT 4.0

· Windows 2000

· Windows .NET

· Windows XP

· Novell NetWare 5.1, 6.0

· Red Hat Linux 7.2, 7.3

· DOS version 6.xx or later

The DOS drivers for MegaRAID are co ntained in the firm ware on MegaRAID, except for the DOS ASPI and CD dr ivers. Call LSI Logic Tec hnical Suppor t at 678-728-1250 or access the web site at

www.lsilogic.com

systems.

for information about drivers for other operatin g

Software Utilities 4-5

4.9 MegaRAID SCSI 320-0 Specifications

Table 4.6 lists the specifications for the SCSI 320-0.

Table 4.6 MegaRAID SCSI 320-0 Specifications

Parameter Specification

Card size 2.536 x 6.6 inches (low-profile PCI) Processor Intel GC80302 integrated I/O processor Bus type PCI 2.2 PCI controller Intel GC80302 Bus data transfer rate Up to 132 Mbytes/s BIOS MegaRAID BIOS Cache configuration Predefined during manufacturing; ECC through a

Firmware 1 MB × 8 flash ROM Nonvolatile RAM 32 Kbytes × 8 for storing RAID configuration Operating voltage 5.00 V ± 0.25 V SCSI data transfer rate Up to 320 Mbytes/s SCSI device types

supported RAID levels supported 0, 1, 5, 10, and 50

4.9.1 PCI Bridge/ CPU

MegaRAID SCSI 320-0 us es the Intel i960RM PCI br idge with an embedded 80960JX RISC proce ssor r unning at 66 MHz. T he RM br idge handles data transfers between the pr imary (host) PCI bus, the secondar y PCI bus, cache memory, and the SCSI bus. The DMA controller suppor ts chaining and unaligned data transfers. The embedded 80960JX CP U directs all controller func tions, including command processi ng, SCSI bus transfers, RAID processing, dr ive rebuilding, cache managemen t, and error recovery.

66MHz 72-bit unbuffered 3.3V SDRAM.

Synchronous or asynchronous. Disk and non-disk.

4-6 Features

4.9.2 Cache Memory

MegaRAID SCSI 320-0 su pports 32 to 12 8 Mbytes of ECC SDRAM DIMM on-board cache mem ory. MegaRAID suppor ts write-through o r write-back caching, wh ich can be selected for each l ogical drive. To improve performance in seque ntial disk accesses, Me gaRAID does not use read-ahead cachi ng for the current logical d rive. The default setting for the read policy is Normal, meaning no read-ahead ca ching. You can disable read-ahead cachi ng.

Warning:

4.9.3 MegaRAID BIOS

The BIOS resides on a 1 Mbyte × 8 flash ROM for easy upgrade. The MegaRAID BIOS suppor ts INT 13h calls to boo t DOS without special software or device drivers. The MegaR AID BIOS provides an extensive setup utility that c an be accessed by pressing <Ctrl><M> at B IOS initialization. The Me gaRAID BIOS Configurati on Utility is descri bed in the MegaRAID Configurati on Software Guide.

4.9.4 SCSI Firmw are

The MegaRAID SCSI 3 20-0 firmware handle s all RAID and SCSI command processi ng and also suppor ts the items in Table 4.7.

Table 4.7 SC SI Firmware

Feature Description

Disconnect/ reconnect

Write caching is not recommended for the physical drives. When write cac he is ena bled, loss of data can oc cur when power is interrupted.

Optimizes SCSI bus seek.

Stripe si ze Variable for all logic al drives: 2, 4 , 8, 16, 32 , 64, or 128

Rebuild Multiple rebuilds and consistency checks with user-

MegaRAID SCSI 320-0 Specifications 4-7

Kbytes.

definable priority.

4.10 RAID Management

RAID management is p rovided by software utilities that manage and configure the RAID syst em and MegaRAID SCSI 320- 0, create and manage multiple disk arrays, control and mon itor multiple RAID se rvers, provide error statisti cs logging, and provide on line maintenance. They include:

· MegaRAID BIOS Configuratio n Utility

· WebBIOS Configuration Utility

· Power Console Plus

· MegaRAID Manager

4.10.1 MegaRAID BIOS Configuration Utility

BIOS Configuration Utility is used to configure and maintain RAID arrays, format hard drives, and manage the RAID system. It is inde pendent of any operating system. See t he MegaRAID Configuration So ftware G u ide for additional information.

4.10.2 WebBIOS Configuration Utility

The WebBIOS Configuration Utility is an HTML-based utility used to configure and maintain RAID arrays, format hard drives, and manage the RAID system. See the Me gaRAID Configuration Software Guide for additional informa tion.

4.10.3 Power Console Plus

Power Console Plus runs in Windows NT, 2000, XP, and .NET. It configures, monitors, and mai ntains multiple RAID ser vers from any network node or a remo te location. See the MegaRA ID Configuration Software Guide for additional informati on.

4.10.4 MegaRAID Manager

This is a character-based ut ility that wor ks in DOS, Novell NetWare, and Red Hat Linux. See the MegaRA ID Configuration Software Guide for additional informa tion.

4-8 Features

4.11 Compatibility

MegaRAID SCSI 320-0 compatibility issues include:

· Server management

· SCSI device compa tibility

· Software compatibility

4.11.1 Server Management

As a simple networ k management protocol (SNM P) agent, MegaRAID SCSI 320-0 suppor ts all SNMP manager s.

4.11.2 SCSI Device Compatibility

MegaRAID SCSI 320-0 suppo r ts SCSI hard dr ives, CD drives, and tape drives.

4.11.3 Software

All SCSI backup and utility software should work with MegaRAID SCSI 320-0. This software is not provid ed with MegaRAID SCSI 320-0 .

Compatibility 4-9

4-10 Features

Chapter 5 Configuring Physical Drives, Arrays, and Logical Drives

This chapter explains how to configure SCS I physical dr ives, arrays, and logical drives. It contai ns the following sections:

· Section 5.1, “Confi guring SCSI Physical Dr ives”

· Section 5.2, “Confi guring Arrays”

· Section 5.3, “Cre ating Logical Drives”

· Section 5.4, “Confi guring Logical Dr ives”

· Section 5.5, “Pla nning the Array Configuration”

5.1 Confi guring SCSI Physical Drives

Physical SCSI drives must be or ganized into logical dr ives. The arrays and logical dri ves that you construct must be able to suppor t the RAID level that you select. The MegaRAID SCSI 320 -0 controller does not have a SCSI channel; the number of SCSI channels depends on the SCSI controller on the m otherboard.

5.1.1 Basic Configuration Rules

You should observe the following guideli nes when connecting and configuring SCS I devices in a RAID array:

· You can pla ce up to 15 physical dr ives in an a rray, depending on the

RAID level.

· Include all drives that h ave the same capacity in the same array.

· Make sure any hot spare has a ca pacity that is at leas t as large as

the largest dri ve that may be replaced by the hot spare.

· When replacing a failed drive, make sure that the re placement dr ive

has a capacity that i s at least as large as the d rive being replaced.

MegaRAID SCSI 320-0 Hardware Guide 5-1

Note: Be su re to back up your data regular ly, even when using

RAID.

5.1.2 Current Physical Device Configuration

Use Table 5.1 to describe the c urrent configuration for your physical devices.

Table 5.1 Physical Device Configur ation

SCSI ID Device Description Termination?

0 1 2 3 4 5

10 11 12 13 14 15

6 8 9

5-2 Configuring Physical Drives, Arrays, and Logical Drives

5.1.3 Logical Drive Configuration

Use Table 5.2 to record the configu ration for your logical drives.

Table 5.2 Logical Drive Configuration

Logical

Drive

LD0 LD1 LD2 LD3 LD4 LD5 LD6 LD7 LD8

LD9 LD10 LD11 LD12 LD13

RAID Level

Stripe

Size

Logical Drive

Size

Cache Policy

Read

Policy

Write

Policy

# of Physical

Drives

LD14 LD15 LD16 LD17 LD18 LD19 LD20 LD21

Configuring SCSI Physical Drives 5-3

Table 5.2 Logical Drive Configuration (Cont.)

Logical

Drive

LD22 LD23 LD24 LD25 LD26 LD27 LD28 LD29 LD30 LD31 LD32 LD33 LD34 LD35

RAID Level

Stripe

Size

Logical Drive

Size

Cache Policy

Read

Policy

Write

Policy

# of Physical

Drives

LD36 LD37 LD38 LD39

5-4 Configuring Physical Drives, Arrays, and Logical Drives

5.1.4 Physical Device Layout

Use Table 5.3 to record the physical device layout.

Table 5.3 Physical Device Layout

Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number

Channel 0 Channel 1

Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number

Configuring SCSI Physical Drives 5-5

Table 5.3 Physical Device Layout (Cont.)

Channel 0 Channel 1

Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number

5-6 Configuring Physical Drives, Arrays, and Logical Drives

Table 5.3 Physical Device Layout (Cont.)

Channel 0 Channel 1

Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level

Configuring SCSI Physical Drives 5-7

Table 5.3 Physical Device Layout (Cont.)

Target ID Device type Logical drive number/Drive number Manufacturer/M odel number Firmware level

5.2 Configur ing Arrays

You organize the physical dis k drives in arrays after they are connected to the SCSI chann els controlled by the LSI Log ic SCSI chip on the motherboard, and after th ey are formatted. An array can consist of u p to 15 physical disk drives, depending on the RAID level.

Channel 0 Channel 1

The MegaRAID SCSI 320-0 su ppor ts up to eight arrays. The number of drives in an array determine s the RAID levels that can be suppor ted.

5.2.1 Arranging Arrays

You must arran ge the arrays to provide additional organ ization for the drive array. You must arrange arrays so that you can create system drives that can function as boot devices.

You can se quentially arrange arrays with an id entical number of dri ves so that the drives in th e group are spanned. Spanne d drives can be treated as one large d rive. Data can b e str iped a cros s multiple arrays as one logical dri ve.

You can cr eate spanned drives by using the Meg aRAID BIOS Setup utility or the MegaRAID Ma nager.

5.2.2 Creating Hot Spares

Any drive that is present, formatted, an d initiali zed, but is not included in a array or logical drive is automa tically designated as a hot spare.

5-8 Configuring Physical Drives, Arrays, and Logical Drives

You can des ignate drives as hot spares u sing the MegaRAID BIOS Configuration Utility, the MegaRAID Manager, or Power Console Plus.

5.3 Creating Logical Drives

Logical drives are arrays or spanned arrays that are presented to the operating system. You must create one or m ore logical drives.

The logical dri ve capacity can i nclude all or any por tion of an array. The logical drive capacity can also be larger than an array by using spanning. The MegaRAID SCSI 3 20-0 supports up to 40 logical dr ives.

5.3.1 Configuration Strategies

The most impor tant factors in RAID array configuration ar e: drive capacity, drive availability (fault tolerance), an d drive performance. You cannot configure a l ogical drive that optim izes all three factors, but it is easy to choose a lo gica l dr ive conf iguration t hat maximizes one factor a t the expense of the other two factors, although n eeds are seldom that simple.

5.3.1.1 Maximize Capacity

RAID 0 achieves maximum drive capacity, but does not provide data redundancy. Maximum drive capacity for each RAID level is shown below. OEM-level firmware tha t can span up to 4 logic al drives is assumed. Table 5.4 descr ibes the RAID levels, including the number of drives required, and the capacity.

Table 5.4 Capacity for RAID Levels

RAID Level Description

0Striping

without parity 1 Mirroring 2 (Capacity of smallest disk) X (1) 5 Striping with

floating parity

drive

Creating L ogical Dri ves 5-9

Drives Required Capacity

1 – 30 (Number of disks) X capacity of

smallest disk

3 – 30 ( Number of di sks) X (capac ity of

smallest disk) - (capacity of 1 disk)

Table 5.4 Capacity for RAID Levels (Cont.)

RAID Level Description

10 Mirroring and

Striping

50 RAID 5 and

Striping

5.3.1.2 Maximizing Drive Availability

You can max im ize the availability o f data on t he physical dis k dr ive in th e logical array by maximizing t he level of fault tolerance. Table 5.5 describes the levels of fault tolerance for the RAID levels.

Table 5.5 Fault Tolerance for RAID Levels

RAID Level Fault Tolerance Protection

0 No fault tolerance. 1 100% protection through data mirroring. 5 100% protection through striping and parity. The data is striped

and parity data is written across a number of physical disk drives.

Drives Required Capacity

4 – 18 (Must be a multiple of 2.)

6 – 30 (Must be a multiple of arrays.)

(Number of disks) X (cap acity of smallest disk ) / (2)

(Number of disks) X (cap acity of smallest dis k) – (c apac ity of 1 d isk X number of Arrays)

10 100% protection through data mirroring. 50 100% protection through data striping and parity. All data is

striped and parity data is written across all drives in two or more arrays.

5.3.1.3 Maximizing Drive Performance

You can co nfigure an array for optimal performance. But optimal dr ive configuration for one type of application will probably no t be optimal for any other application. A basic guideline of t he performance characteristic s for RAID drive arrays at each RAID level is shown in

Table 5.6.

5-10 Configuring Physical Drives, Arrays, and Logical Drives

Table 5.6 Performance Characteristics for RAID Levels

RAID Level Performance Characteristics

0 Excellent for all types of I/O activity, but provides no data

1 Provides data redundancy and good performance. 5 Provides data redundancy and good performance in most

10 Provides data redundancy and excellent performance. 50 Provides data redundancy and very good performance.

security.

environments.

5.3.2 Assigning RAID Levels

Only one RAID level can be assigned to ea ch logical drive. Table 5.7 shows the drives required per RAID level.

Table 5.7 Number of Physical Driv es for RAID Levels

RAID

Level Minimum # of Physical Drives Maximum # of Physical Drives

01 30 12 2

53 30 10 4 18 50 6 30

Note:

Creating L ogical Dri ves 5-11

The MegaRAID SCSI 320-0 controller supports one or two SCSI channels, with up to 15 SCSI devices on each channel.

5.4 Configur ing Logical Drives

After you have installed the MegaRAID SCS I 320-0 controller in the server and have attached all physical disk drives, perform the following actions to prepare a RA ID array:

1. Optimize MegaRAID SCSI 320 -0 controller options for your system. See Chapter 6 for additional i nformation.

2. Press <Ctrl><M> to r un the BIOS Config uration Utility.

3. If necessary, perform a low-level format of the SCSI drives that will be included in the ar ray and the drives to be used for hot spares.

4. Customize the RAID array and define and config ure one or more logical drives by selecti ng Easy Configuration or New Configuration.

5. Create and configure one or mor e system drives (logical drives) by selecting the RAID level, cache policy, read policy, and write policy.

6. Save the configuration.

7. Initialize the system dr ives. After initializati on, you can install the operatin g system.

5.4.1 Optimizing Data Storage

5.4.1.1 Data Access Requirements

Each type of data stored in the disk subs ystem has a different freque ncy of read and writ e acti vity. If you know the data access requirements, you can more successf ully determine a strategy for optimizing the disk subsystem capaci ty, availability, and performance.

Servers that support Video on Dema nd typica lly read the da ta often, but write data infrequ ently. Both the read and write operations tend to be long. Data stored on a general-purpos e file server involves relatively short rea d and write operation s with relatively small files.

5.4.1.2 Array Functions

You must fir st define the major pur pose of the disk ar ray. Will this disk array increase the system storage ca pacity for general-purp ose file and

5-12 Configuring Physical Drives, Arrays, and Logical Drives

print servers? Doe s this disk array suppor t any software system that must be available 24 hours per day? Will the informa tion stored in this disk array contain large audi o or video files that must be available on demand? Will this dis k array contain data from a n imaging s ystem? You must identify the p urpo se of th e data to be sto red in the disk sub system before you can confidently choose a RAID level and a RAID configuration.

5.5 Planni ng the Array Configuration

Fill out Table 5.8 to help plan this array.

Table 5.8 Factors for Planning the Array Configuration

Question Answer

Number of physical disk drives in the array Purpose of this array. Rank the following factors:

Maximize drive cap acity Maximize the safety of the data (fault tolerance)

Maximize hard drive performance and throughput Number of hot spares Amount of cache memory installed on MegaRAID SCSI 320-0 Are all of the disk drives and the server protected by a UPS?

5.5.1 Using the Array Configuration Planner

The following table lists the possi ble RAID levels, fault tolerance, and effective capacity for all possible drive configuratio ns for an array consisting of one to e ight drives. This table does not take into account any hot spare (standby) drives. You should always have a hot spare drive in case of drive failure. RAID 1 requires two drives, RAID 10 at least four, and RAID 50 at least six.

Planning the Array Configuration 5-13

Use Table 5.9 to plan the array configuration.

Table 5.9 Array Configuration Planner

# of Drives

Possible

RAID Levels

Relative

Performance

Fault

Tolerance

1 None Excellent No 100% 1 RAID 0 Excellent No 100% 2 None Excellent No 100% 2 RAID 0 Excellent No 100% 2 RAID 1 Good Yes 50% 3 None Excellent No 100% 3 RAID 0 Excellent No 100% 3 RAID 5 Good Yes 67% 4 None Excellent No 100% 4 RAID 0 Excellent No 100% 4 RAID 5 Good Yes 75% 4 RAID 10 Good Yes 50% 5 None Excellent No 100% 5 RAID 0 Excellent No 100%

Effective Capacity

5 RAID 5 Good Yes 80% 6 None Excellent No 100% 6 RAID 0 Excellent No 100% 6 RAID 5 Good Yes 83% 6 RAID 10 Good Yes 50% 6 RAID 50 Good Yes 67% 7 None Excellent No 100% 7 RAID 0 Excellent No 100% 7 RAID 5 Good Yes 86% 8 None Excellent No 100%

5-14 Configuring Physical Drives, Arrays, and Logical Drives

Table 5.9 Array Configuration Planner (Cont.)

# of Drives

Possible

RAID Levels

Relative

Performance

Fault

Tolerance

8 RAID 0 Excellent No 100% 8 RAID 5 Good Yes 87% 8 RAID 10 Good Yes 50% 8 RAID 50 Good Yes 75%

Effective Capacity

Planning the Array Configuration 5-15

5-16 Configuring Physical Drives, Arrays, and Logical Drives

Chapter 6 Hardware Installation

This chapter explains how to install the MegaRAID SCSI 320-0 controller. It contains the following secti ons:

· Section 6.1, “Hardware Re quirements”

· Section 6.2, “Instal lation Steps”

· Section 6.3, “Summar y”

6.1 Hardware Requirements

You must have the following in order to install the MegaRAID SCSI 320-0 controller and creat e arrays:

· A host computer with the following:

– Special ZCR P CI expansion slot – Installed LS I Logic 53C1030 SCSI c hip – Suppor t for PCI version 2.2 or later – Intel Pentium, Pentium Pro, or more powerful CPU – Floppy drive, color monitor, VGA adapter card , mouse, and

keyboard

· The MegaRAID SCSI 3 20-0 Installation CD

· An Uninterruptible Power Supply (UPS) for the en tire system

· Ultra320, Ultra, Fast SCSI 2 or Wide SCSI har d disk drives

MegaRAID SCSI 320-0 Hardware Guide 6-1

6.2 Installation Steps

The MegaRAID SCSI 3 20-0 provides extensive customization option s. If you need only basic MegaRA ID SCS I 320-0 features and your compu ter does not use other adapter cards with resource settings that may conflict with MegaRAID SCSI 32 0-0 settings, even custom installation can be quick and easy.

Table 6.1 lists the hardware in stallation st eps. Each step is d escribed in

detail in the following pages.

Table 6.1 Hardware Installation Steps

Step Action Additional Information

1 Unpack the MegaRAID controller and

2 Turn the computer off and remove the

3 Check the jumper settings on the

4 Install the MegaRAID SCSI 320-0 card. 5 Replace the computer cover and turn the

6 Run MegaRAID BIOS Configuration Utility. Optional. 7 Install software dr ivers for the de sired

6.2.1 Step 1: Unpack

Unpack and install the hardware i n a static-free environment. R emove the MegaRAID SCSI 320-0 controller card from the anti-static bag and inspect it for damage. If the c ard appears damaged, o r if any item lis ted below is missing, conta ct LSI Logic or your MegaRAI D OEM suppor t representative. The MegaRAID SCSI 320- 0 co ntrol ler is shi pp ed wit h the following:

inspect for dama ge. Make sure all i tems are in the package.

cover.

MegaRAID SCSI 320-0 controller.

power on.

operating systems.

If damaged, call your LSI Logic OEM support representative.

See page 6-3for the MegaRAID SCSI 320-0 jumper settings.

Be sure the S CSI devices are powered up before or at the same time as the host computer.

6-2 Hardware Installation

· The Driver and Documenta tion CD, which contains these items:

– The M egaRAID Configuration Soft ware Guide – The M egaRAID Operating Sys tem Driver Installation Gu ide – The M egaRAID SCSI 320-0 ZCR Hardware Guide – The software l icense agreement – The Mega RAID SCSI 320-0 config uration utilities for DOS

· The warranty registration car d

6.2.2 Step 2: Power Down

Tur n off the computer, remove the power cord, then remove the cover. Make sure the computer is tur ned off and disconne cted from any networks before installi ng the controller card.

6.2.3 Step 3: Set Jumpers

Make sure the jumper settin gs on the MegaRAID SCSI 32 0-0 card are correct. The jumpers are set at the factory and you probably do not nee d to change them. Table 6.2 lists th e jumpers (and a co nnector).

Table 6.2 Jumpers for the MegaRAID SCSI 320-0

Item Description Type

J2 Dirty cache LED 2-pin header J3 Clear EPROM 2-pin header J4 BIOS enable 2-pin header J8 BBU daughter card connector 40-pin connector

Figure 6.1 shows the location of th ese items on the MegaRA ID SCSI

320-0 controller.

Installat ion Steps 6-3

Figure 6.1 MegaRAID SCSI 320-0 Controller Layout

6.2.3.1 J2 Dirty Cache LED

J2 is a two-pin heade r for the dirty cache L ED. This can be connec ted to an LED on the com puter enclosure. The LED will be lit when data in the cache has not yet bee n written to the storage d evice.

J2 J3 J4

Optional Backup

Battery Unit

Connector

J8

6.2.3.2 J3 Clear EPROM

J3 is a two-pin header th at clears the erasable progammable read-on ly memory (EPROM) c onfiguration data.

6.2.3.3 J4 BIOS Enable

J4 is a 2-pin header that enables or disables the Mega RAID onboard BIOS. The onboard BIOS sh oul d be ena bled (J4 un jum per ed ) for nor mal board position. Table 6.3 conta ins the pinout for J4.

Table 6.3 Pinout for J4 BIOS Enable

J4 Setting Onboard BIOS Status

Unjumpered Enabled Jumpered Disabled

6-4 Hardware Installation

6.2.3.4 J8 Connector for Optional BBU Daughter Card

J8 is 40-pin conn ector that is use d to mount an opti onal batter y backup unit that is on a daughte rcard.

6.2.4 Step 4: Install MegaRAID SCSI 320-0

Select the ZCR 3.3 V or 5 V PCI slot and align the MegaRAID SCSI 320-0 controller bus connector wi th the slot. Figure 6.2 shows the differences between the 3.3 V and 5 V slots.

Figure 6.2 3.3 V and 5V Slots

Note:

The slot must be a specific ZCR slot on a motherboard that has an installed LSI Logic 53C1030 SCSI c hip; see your motherboard manual for this in formation.

Insert the Meg aRAID SCSI 320- 0 card in the ZCR PCI slot, as s hown in

Figure 6.3. Press down gently but fir mly to make sure that the card is

properly seated in the slot. The bottom edge of the controller card should be flush with the s lot. Attach t he bracket to the compu ter frame with the bracket screw.

Installat ion Steps 6-5

Figure 6.3 Installing the MegaRAID SCSI 320-0 Controller

Bracket Screw

6.2.5 Step 5: Power Up

Replace the computer cover and reconnect the AC power cords. Tu rn power on to the host computer. Set up the power suppli es so that the SCSI devices are powered up at the s ame time as or before the host computer. If the computer is powered up be fore a SCSI device, the device might not be recognized.

During boot, the Meg aRAID SCSI 320-0 BIOS mess age appears:

MegaRAID SCSI 320-0 Disk Array Adap ter BIOS Version x.xx date

Copyright (c) LSI Logic Corporation

Firmware Initializing... [ Scanning SCSI Device ...(etc.)... ]

32-Bit Slots (3.3 V)

64-Bit Slots (5 V)

6-6 Hardware Installation

The firmware takes several seconds to initial ize. During this time the adapter will scan the SC SI channel(s). When r eady, the following appears:

Host Adapter-1 Fi rmware Version x.xx DRAM Size 16 MB 0 Logical Drives fo und on the Host Adapter 0 Logical Drives ha ndled by BIOS Press <Ctrl>< M> to run M egaRAID SCS I 320-0 B IOS Configuration Utility

The <Ctrl><M> u tility prompt times out a fter several seconds. The MegaRAID SCSI 320-0 hos t adapter (controller) numbe r, firmware version, and cache DRAM size are displayed in the second portion of the BIOS message. The number ing of the controllers follows the PC I slot scanning order used by the host motherboard.

6.2.6 Step 6: Run the MegaRAID BIOS Configuration Utility

Press <Ctrl><M> to run the MegaRAID BIOS Configuration Utility. See the MegaRAID Configurati on Software Guide for information ab out running this uti lity.

6.2.7 Step 7: I nstall the Operat ing System Driver

MegaRAID can operate unde r MS-DOS or any DOS-compatible operating system using the s tandard AT BIOS INT 13h Hard Disk Drive interface. To op erate with other operating syste ms, you must install software drivers. MegaRAID provides software dr ivers on the Driver and Documentation CD for the following operating s ystems:

· MS-DOS version 6.xx or later

· Microsoft Windows NT 4.0, Windows 2000, Windows XP , Windows

.NET

· Novell NetWare 5.1, 6.0

· Red Hat Linux 7.2, 7.3

Note:

Important:

Refer to the MegaRAID Driver Installation Gu ide for the procedures used to install operating system dr ivers.

When booting the sy stem from a drive connected to a MegaRAID controller and using EMM386.EXE, MEGASPI.SYS must be loaded in CONFIG.SYS before

Installat ion Steps 6-7

6.3 Summary

This chapter discussed hardware installation. Configure the RAID system using software config uration utilities. See the Mega RAID Configuration Software Guide for all information about MegaRAID SCSI 320-0 software utilities. The utility programs for configuring MegaRAID SC SI 320-0 are shown in Table 6.4.

Table 6.4 Configuration Utilities and Opera ting Systems

EMM386.EXE is loade d. If you do not do this, you cannot access the boot dr ive after EMM386 is loaded.

Configuration Utility Operating System

MegaRAID BIOS Configuration Utility Independent of the operating system WebBIOS Configuration Utility Independent of the operating system MegaRAID Manager DOS

Power Console Plus Microsoft Windows NT

Red Hat Linux 7.2, 7.3 Novell NetWare 5.1, 6.0

Windows 2000 Windows XP Windows .NET

6-8 Hardware Installation

Chapter 7 Troubleshooting

This chapter provides tr oubleshooting information for the MegaRA ID SCSI 320-0 controll er. It contains the following sections:

· Section 7.1, “General Troublesh ooting”

· Section 7.2, “BIOS Boot Error Messages”

· Section 7.3, “Other BIOS Error Messages”

· Section 7.4, “Other Potential Problems”

7.1 General Troubleshooting

This section lists the general problems that can occ ur, along with suggested soluti ons. Table 7.1 des cribes general pro blems you might encounter, along with suggested solut ions.

Table 7.1 General Problems and Suggested Solutions

Problem Suggested Solution

The system hangs during the boot process after installation.

Some operating syste ms do no t loa d in a computer with a MegaRAID SCSI 320-0 adapter.

Make sure the SCSI BIOS on the motherboard has been disabled. Make sure the Me gaRAID SCSI 320-0 adapt er card is installed in the proper PCI expansion slot. It must be installed in the RAID Upgrade PCI slot.

Check the system BIOS configuration for PCI interrupt assignments. Make sure some Interrupts are assigned for PCI. Initialize the logical drive before installing the operating system.

MegaRAID SCSI 320-0 Hardware Guide 7-1

Table 7.1 General Problems and Suggested Solutions (Cont.)

Problem Suggested Solution

One of the hard drives in the array fails often.

Check the driv e error cou nts using Power Console Plus . See the MegaRAID Configuration Software Guide for more information. Format the d rive. Rebuild the drive If the drive continues to fail, replace the drive with another drive with the same capacity.

Pressed <Ctrl><M>. Ran Megaconf.exe and tried to make a new

Check the drives IDs on each channel to make sure each

device has a different ID. configurati on. The sy stem ha ngs when scanning devices.

Check the termination. The device at the end of the channel

must be ter minated.

Replace the drive cable. Multiple drives using the same power

supply. There is a problem spinning the

Set the drives to spin on command. This will allow

MegaRAID SCSI 320-0 to spin two devices simultaneously. drives all at once.

Pressing <Ctrl><M> or running

These util ities requi re a color monitor. megaconf.exe does not display the Management Menu.

Cannot flash or update the EEPROM. You may need a new EEPROM.

Firmware Initia lizing...

Make sure that TERMPWR is being properly provided to

each peripheral device populated channel. appears and remains on the screen.

Make sure that each end of the channel chain is properly

terminated using the recommended terminator type for the

peripheral device.

What SCSI IDs can a non-hard disk device have and what is maximum number allowed per adapter?

7-2 Troubleshooting

Make sure (on a channel basis) only two types of cables are

connected at any one time.

Make sure that the MegaRAID SCSI 320-0 controller is

properly seated in the ZCR PCI slot.

Non-hard disk devices can accommodate only SCSI IDs 1,

2, 3, 4, 5 or 6, regardless of the channel used. A maximum

of six non-hard disk devices are supported per MegaRAID

SCSI 320-0 adapter.

7.2 BIOS Boot Error Messages

Table 7.2 describes err or messages tha t can display when you boot the

system, and suggest ed solutions.

Table 7.2 BIOS Boot Error Mess ages

Message Problem Suggested Solution

Adapter BI OS Disabled . No Logical Drives Handled by BIOS

Host Adapt er at Basepor t xxxx Not Responding

No MegaRAID SCSI 320-0 Adapter

Configura tion of nonvolatile RAM (NVRAM) and dr ives mismatch. Run View/Add Configura tion option of Configuration Utility. Press any key to run the Configuration Utility.

1 Logical Drive Failed A logical drive failed to

The MegaRAID BIOS is disabled. Som etime s the BIOS is disabled to prevent booting from the BIOS.

The BIOS cannot communicate with the adapter firmware.

The configur ati on sto red in the MegaRAID SCSI 320-0 adapter does not match the configuration stored in the drives.

sign on.

Enable the BIOS using the MegaRAID BIOS Configuration Utility.

Make sure MegaR AID SCSI 320 -0 is proper ly installed.

1. Press a key to run MegaRAID Manager.

2. Select View/Add Configuration from the Configure menu.

3. Use View/Add Configuration to examine both the configuration in NVRAM and the configuration stored on the disk drives.

4. Resolve the problem by selecting one of the configurations.

Make sure all physical drives are properly connected and are powered on.

Run MegaRAID Manager to find out if any physical drives are not responding. Reconnect, replace, or rebuild any dr ive that is not responding.

X Logical Drives Degraded

x number of l ogical drives signed on in a degraded sta te.

BIOS Boot Error Messages 7-3

Make sure all physical drives are properly connected and are powered on.

Run MegaRAID Manager to find if any physical drives are not responding. Reconnect, replace, or rebuild any drive that is not respon ding.

Table 7.2 BIOS Boot Error Mess ages (Cont.)

Message Problem Suggested Solution

1 Logical Drive Degraded

The following SCSI IDs are not responding: Channel x:a.b.c

A logical drive signed on in a degraded state.

The physical drives with SCSI IDs a, b, and c are not responding on SCSI channel x.

Make sure all physical drives are properly connected and are powered on.

Run MegaRAID Manager to find out if any physical drives are not responding. Reconnect, replace, or rebuild any dr ive that is not responding.

Make sure the physical drives are properly connected and are powered on.

7.3 Other BIOS Error Messages

Table 7.3 describes other BI OS error messages, their meani ng, and

suggested soluti ons.

Table 7.3 Other BIOS Error Messages

Message Problem Suggested Solution

Following SCSI disk not found and no empty slot available for mapping it

The physical disk roaming feature did not find the physical disk with the displayed SCSI ID. No slot is available to map the physical drive. MegaRAID cannot resolve the physical drives into the current configuration.

Reconfigure the array.

Following SCSI IDs have the same data y, z Channel x: a, b, c

7-4 Troubleshooting

The physical dr ive roaming feature found the same data on two or more physical drive on channel x with SCSI IDs a, b, and c. MegaRAID cannot determine the drive that has the duplicate information.

Remove the drive or drives that should not be used.

Table 7.3 Other BIOS Error Messages (Cont.)

Message Problem Suggested Solution

Unresolved configuration mismatch between disks and NVRAM on the adapter

The configuration stored in the MegaRAID NVRAM does not match the configur ation stored on the drives.

1. Press a key to run MegaRAID Manager.

2. Select View/Add Configuration from the Configure menu.

3. Use View/ Add Con figurati on to examin e both the configura tion in N VRAM and the configuration stored on the disk drives.

4. Resolve the problem by selecting one of the configurations.

7.4 Other Potential Problems

Table 7.4 describes other po tential problems.

Table 7.4 Other Potential Problems

Topic Information

DOS ASPI MEGASPI.SYS, the MegaRAID DOS ASPI manager, uses 6 Kbytes of system

CD drives under DOS

Physical drive errors To display the MegaRAID Manager Media Error and Other Error options, select

memory once it is loaded. At this time, copied CDs are not accessible from DOS even after loading

MEGASPI.SYS and MEGACDR.SYS.

the Objects menu, then Physical Drive. Select a physical drive and press <F2>. The windows di splays the number of errors. A Media Error is an error that occurred while actually transferring data. An Other Error is an error that occurs at the hardware level because of a device failure, poor cabling, bad termination, signal loss, etc.

Virtual sizing The virtual sizing option enables RAID expansion. Virtual sizing must be

enabled to increase the size of a logical drive or add a physical drive to an existing logical drive. Run the BIOS Configuration Utility by pressing <Ctrl><M> to enable virtual sizing. Select Objects, then Logical Drive. Next, select View/Update Parameters, then set vi rtual sizing to Ena bled.

BSD Unix We do not provide a driver for BSDI Unix. MegaRAID SCSI 320-0 does not

support BSDI Unix.

Multiple LUNs MegaRA ID SC SI 3 20-0 su ppo rts one logi cal unit number (LUN ) pe r ea ch targ et

ID. No multiple LUN devices are supported.

Other Potential Proble ms 7-5

Table 7.4 Other Potential Problems (Cont.)

Topic Information

MegaRAID power requirements

SCSI bus requirements

Windows NT Installation

The maximum MegaRAID SCSI 320-0 power requirements are 15 W at 5 V and 3 A.

The ANSI specification dictates the following:

· The maximum signal path length between terminators is 3 meters when

using up to 4 maximum capacitance (25 pF) devices and 1.5 meters when using more than 4 devices.

· SCSI devices should be uniformly spaced between terminato rs, with the end

devices located as close as possible to the terminators.

· The characteristic impedance of the cable should be 90 +/- 6 ohms for the

/REQ and /ACK signals and 90 +/- 10 ohms for all other signals.

· The stub length (the distance from the controller's external connector to the

mainline SCSI bus) shall not exceed 0.1m (approximately 4 inches).

· The spacing of devices on the mainline SCSI bus should be at least three

times the stub length.

· All signal lines shall be terminated once at both ends of the bus powered

by the TERMPWR line.

When Windows NT is installed using a bootable CD, the devices on the MegaRAID SCSI 320-0 will not be recognized until after the initial reboot. The Microsoft documented workaround is in SETUP.TXT. SETUP.TXT is on the CD. To install drivers when Setup recognizes one of the supported SCSI host adapters without making the devices attached to it available for use:

1. Restart Windows NT Setup.

2. Press <F6> to prevent Windows NT Setup from performing disk controller detection when Windows NT Setup displays the following:

Setup is inspecting your computer's hardware configuration...,

This allows you to install t he driver fr om the dr ivers disk you crea ted. All SCSI adapters must be installed manually.

3. Press <S> to display a list of supported SCSI host adapters when Windows NT Setup di splays the followin g:

Setup could not determine the type of one or more mass storage devices installed in your system, or you have chosen to ma nually speci fy an adapter,

4. Select Other from the bottom of the list.

5. Insert the drivers disk you made when prompted to do so and select MegaRAID SCSI 320-0 from this list. In some cases, Windows NT Setup repeatedly prompts to swap disks. Windows NT will recognize any devices attached to this adapter.

6. Repeat this step for each host adapter not already recognized by Windows NT Setup.

7-6 Troubleshooting

Appendix A Audible Warnings

The MegaRAID SCSI 3 20-0 RAID controller has a n onboard tone generator that indicate s events and errors.

Note: This i s available only if the optional seri es 502 Battery

Backup Unit (BBU) is ins talled.

Table A.1 Audible Warnings and Descriptions

Tone Pattern Meaning Examples

Three seconds on and one second off

One second on and one second off

One second on and three seconds off

A logical drive is offline.

A logical drive is running in degraded mode.

An automatically initiated rebuild h as b een completed.

One or more drives in a RAID 0 configuration failed.

Two or more dri ves i n a RA ID 1, or 5 configuration failed.

One drive in a RAID 5 configuration failed.

Whil e you w ere away from the system, a disk drive in a RAID 1, or 5 configuration f ail ed and was rebuilt.

MegaRAID SCSI 320-0 Hardware Guide A-1

A-2 Audible Warnings

Appendix B Glossary

Array A grouping of individual di sk drives that combines the storage space on

the disk drives into a s ingle segment of co ntiguous storage space. MegaRAID can group disk dr ives on one or more SCSI c hannels into an array.

Array Management Software

Array Spanning Array spanning by a logical drive combines storage space i n two arrays

Asynchronous Operations

Cache I/O A small am ount of fast memory that hol ds recently accessed data.

Software that provides common control and management for a disk array. Array management software most oft en executes in a disk controller or intelligent host bus adapte r, but it can also execute in a host compute r. When array management software executes in a disk contr oller or adapter, it is often called firmware.

of disk drives into a si ngle, contiguous storage spac e in a logical dr ive. MegaRAID logical d rives can span consecuti vely numbered arrays that have the same number of disk dri ves. Array spanning promotes RAID levels 1 and 5 to RAID levels 10 and 50, respectively. See also Disk Spanning.

Operations that bear n o relationship to each ot her in time and can overlap. The concept of asynchrono us I/O operations is central to independent access arrays in throughput-intensive applicati ons.

Caching speeds subseq uent access to the same d ata. It is most often applied to processor-memory access, but it can also be used to store a copy of data accessible over a network. Wh en data is read from or written to main memory, a copy is also saved in cache memory with the associated main memory address. The cache memory software monitors the addresses of s ubse quent r eads to se e if t he re quired d ata is alr eady stored in cache mem ory. If it is already in cache memory (a cache h it), the data is read from cach e memory immediately and th e main memor y read is abor ted (or not sta rted.) If the data is not cached (a ca che mi ss), it is fetched from main me mory and saved in cache me mory.

MegaRAID SCSI 320-0 Hardware Guide B-1

Channel An electrical path for the transfer of data and control information between

a disk and a disk contr oller.

Consistency Check

An examination of the disk system to determ ine whether all conditio ns are valid for the specified configuratio n (such as parity.)

Cold Swap A cold swap requires that you turn the power off before replacing a

defective disk drive in a disk subsyste m.

Data Transfer Capacity

The amount of data per unit time moved through a channe l. For disk I/O, data transfer capacity (bandwidth) is expressed in megabytes per second (Mbytes/s).

Degraded Used to desc ribe a disk drive that h as become non-functional o r has

decreased in performa nce.

Disk A nonvolatile, randomly addressable, rewritable mass storage device,

including both rotating mag netic and optical disks and solid- state disks, or nonvolatile electronic storage elements. It does not include specialized devices such as write-once-rea d-many (WORM) optical disks, nor does it include so-called RAM disks implemented using software to control a dedicated por tion of a host co mputer volatile random acce ss memory.

Disk Array A collection of disks from one or mo re disk subsystems combi ned with

array management software. The software cont rols the disks and presents them to the array operating environment as on e or mo re v ir tu al disks.

Disk Duplexing A variation on d isk mirroring in whic h a second disk adapter or host

adapter and redundant disk drives are present.

Disk Mirroring Writing duplic ate data to more than one (u sually two) disk drives to

protect against data loss in the event of device failure. Disk mirror ing is a common feature of RAID systems.

Disk Spanning Disk spanning allows mult iple disk drives to functi on like one big drive.

Spanning overcomes lack of disk s pace and simplifies storag e management by combinin g existing resources or adding rel atively inexpensive resources. For example, four 36 Gbyte disk drives can be combined to appear to t he operating system as one single 144 Gbyte drive. See also Array Spanning and Spanning.

Disk Striping A type of disk array mapping. Con secutive stripes of data are ma pped

round-robin to conse cutive array members. A striped array (RAID 0)

B-2 Glossary

provides high I/O perform ance at low cost, but provides lowers da ta reliability than any of its member disks.

Disk Subsystem A collection of disks and the hardware that connects them to one or more

host computers. The hardware can include an intelligent controller, or the disks can attach di rectly to a host compute r I/O bus adapter.

Double Buffering

A technique that achieves maximum da ta transfer bandwidth by constantly keeping two I/O r equests for adjacent data outstan ding. A software component begins a double-buffered I/O stream by issuing two requests in rapid se quence. Thereafter, each time an I/O req uest completes, another is i mmediately issued. I f the disk subsystem is capable of processing requests fast enough, double buffering allows data to be transferred at the full-volume transfer rate.

Failed Drive A drive that has ceased to function or consisten tly functions improper ly. Fast SCSI A variant on t he SCSI-2 standard. It uses the same 8-bit bus as the

original SCSI-1 , but runs at up to 10 Mbytes (double the speed of SCS I-

1.)

Firmware So ftware stored in read-only me mory (ROM) or Programmable ROM

(PROM). Firmware often controls the behavior of a system when it is first turned on. A ty pical example woul d be a mo nitor program in a comput er that loads the full operati ng sys tem from disk or from a network an d the n passes control to t he operating system.

FlexRAID Po wer Fail Option

The FlexRAID Power Fail option allows a reconstructi on to restart if a power failure occurs. This is the advantage of t his option. The disadvantage is, once the reconstr uction is active, the performance is slower because an addition al activity is added.

Format The process of wr iting zeros to all data fields in a physical drive (disk

drive) to map out unre adable or bad sectors. Bec ause most disk dr ives are factory formatted , formatting is usua lly only done if a har d disk generates many media error s.

GByte Giga byte. Shorthand for 1,000,000,000 (10 to the ninth power) bytes.

One Gbyte is equivalent to 1,00 0 Mbytes.

Host-based Array

A disk array with an array management software in i ts host computer rather than in a disk sub system.

B-3

Host Computer Any comp uter to whi ch disks a re direct ly attach ed. Mainfram es, ser vers,

workstations, and pers onal computers can all be considered hos t computers.

Hot Spare A s tand - by disk drive ready for use if a drive in an array fails. A hot spare

does not contain any user data. Up to eight disk dr ives can be assigned as hot spares for an adapter. A hot spare ca n be dedicated to a sin gle redundant array, or it can be par t of the global hot-spa re pool for all arrays controlled by the adapter.

Hot Swapping The substitution of a replacement unit in a disk subsystem for a defective

one, where the substitution can be performed while th e subsystem is running. Hot swaps are done manuall y.

I/O Driver A host c omputer software component ( usually part of the operating

system) that control s the operation of per ipheral controller s or adapters attached to the host com puter. I/O drivers communicate between applications and I/O devices, and in some ca ses they par t icipate in data transfer.

Initialization The process of writing zeros to the data fields of a logical d rive and

generating the correspondi ng parity to put the log ical drive in a Ready state. Initializing erase s previous data and generates par ity so that the logical drive will pa ss a consistency check. Arrays can wo rk without initialization, but they can fail a con sistency check because the parity fields have not been generated.

Logical Disk A set of contiguous chunks on a physical disk. Log ical disks are us ed in

array implementations as co nstituents of logi cal volumes or par titions. Logical disks ar e normally transp arent to the host environment, except when the array containing them i s being configured.

Logical Drive A virtua l dri ve within a n array that ca n con sist of more than one physical

drive. Logical drives divid e the contiguous storage space of an array of disk drives or a spanned gro up of arrays of drives. The storage spa ce in a logical drive is spr ead across all the physica l drives in the array or spanned arrays. Configure at least o ne logical drive for each array.

Mapping The conversion between multiple data addressing s chemes, especially

conversions between member disk block addresses and block addresses of the virt ual disks presented to the operating environment by array management software.

B-4 Glossary

Mbyte (Megabyte) An abbreviation for 1,000,000 (10 to the sixth power) bytes.

One Mbyte equals 1,000 Kbytes (kilobytes).

Multi-threaded Having mul tiple concurrent or p seudo-concurrent execution sequence s.

Used to describ e processes in computer systems. Multi-threaded processes allow throughput-intensive applications to efficiently use a disk array to increase I/O performa nce.

Operating Environment

The operating environment incl udes the host computer wher e the array is attached, any I/O buses and a dapter s, the host op erating sy stem, and any additional software requ ired to operate the array. For host-based arrays, the operating environment include s I/O driver software for the member disks, but does not include ar ray management software, which is regarded as par t of the array itself.

Parity Parity is an extra bit added to a byte or word to reveal errors in storage

(in RAM or disk) or transmission. Parity is used to generate a set of redundancy data from two or more parent data sets. The redundancy data can be used to recons truct one of the paren t data sets, although it does not fully dupl icate the parent data sets. In RAID, this method is applied to entire dri ves or stripes across all disk drives in an array. Parity data is distributed am ong all the drives in the s ystem. If a single disk drive fails, the drive can be rebuilt from the par ity of the res pective data on the remaining dr ives.

Partition An array virtua l disk made up of logi cal disks rather t han physical ones.

Also known as logical volume.

Physical Disk A hard disk drive that stores data. A hard disk drive consis ts of one or

more rigid magneti c discs rotating about a c entral axle with associated read/write heads and electronics.

Physical Disk Roaming

The ability of some adapters to detect when disk drives have been moved to a different slot in the computer—for example, after a hot swap.

Protocol A set of formal rules desc ribing how to transmit data, e specially across

a network. Low level protocols define the electrical and physica l standards to be obser ved, bit- and byte- ordering, a nd the transmission and error detection and correction of the bi t stream. High level protocols deal with the data formatting, inc lud ing the mes sage syn tax , the terminalto-computer dialogue, chara cter sets, and sequencing o f messages.

B-5

RAID Redunda nt Array of Indep endent Disks. A data sto rage method in which

data, along with parity information, is distributed among two or more hard disks (called an array) to improve performance and reliability. A RAID disk subsystem improves I/O performance on a server using only a single drive. The RAID array appears to the host server as a single storage unit. I/O is expedited because several disks can be acces sed simultaneously.

RAID Levels A style of red undancy applied to a l ogical drive. It can increa se the

performance of the logic al drive and can decrease usa ble capaci ty. Each logical drive must have a RAID level assigned to it. The RAID level drive requirements are: RAID 0 req uires one or more physical drives, RAID 1 requires exactly two physical drives, RAID 5 requir es at least three physical drives. RAID levels 10 and 50 result wh en logical drives span arrays. RAID 10 results when a RAID 1 logical dr ive spans arrays. RAID 50 results when a R AID 5 logical drive spans a rrays.

RAID Migration RAI D migration is used to move between optimal RAID l evels or to

change from a degraded re dundant logical drive to an opt imal RAID 0. In Novell, the utility used for RAID migration is MEGAMGR; in Windows NT, it is Power Console Plus. If a RAID 1 array is being converted to a RAID 0 array, instead of performi ng RAID migration, one drive can be removed and the o ther reconfigured o n the controller as a RAID 0. This is due to the same data being wr itten to each drive.

Read-Ahead A memor y caching capability in some adap ters that all ows them to read

sequentially ahead o f requested data and sto re the additional data in cache memor y, anticipating that the additional dat a will be neede d soon . Read-Ahead supplies s equential data faster, but it is not as effective when accessing random da ta.

Ready State A condition in which a workable disk drive is neither online nor a hot

spare and is available to add to an array or to designate as a hot spare.

Rebuild The regeneration of all d ata from a failed disk in a RAID level 1, or 5

array to a replacement disk. A disk rebuild normally occurs without interruption of applica tion access to data stor ed on the ar ray virtu al disk.

Rebuild Rate The p ercentage of CPU reso urces devoted to rebuilding. Reconstruct The act of remaking a lo gical dr ive after changing RAID l evels or adding

a physical drive to an existing array.

B-6 Glossary

Redundancy T he provision of multiple intercha ngeable components to perform a

single function to co pe with failures or errors. Redundan cy normally applies to hardware; disk mirroring is a com mon form of hardware redundancy.

Replacement

A disk available to replace a failed member disk in a RAID array.

Disk Replacement

Unit

A component or co llection of component s in a disk subsyste m that is always replaced as a unit when any part of the collection fails. Typical replacement units i n a disk subsystem in cludes disks, contro ller logic boards, power supplies, and ca bles. Also called a hot spare.

SAF-TE SCSI-access ed fault-tolerant enclosure. An ind ustry protoco l for

managing RAID enclo sures and repor ting enclosure environmental information.

SCSI (Small computer system interface) A processor-independent standard for

system-level interfacing between a computer and intelligent devices, including hard dis ks, floppy disks, CD-ROM, printers, scanners, etc . Multiple SCSI devices can be c onnected to a single host adapter on the computer's bus. SCSI transfers eight or 1 6 bits in parallel and can operate in either asynchronous or synchronous modes. The synchronous transfer rate is up to 320 Mbytes/s (for Ultra320 SCSI). SCSI connections normally us e single-ended drivers, as oppos ed to differential drivers.

SCSI Channel MegaRAID controls the disk dr ives through SCSI-2 buses (ch annels)

over which the system transfers data in either Fast and Wide or Ultra SCSI mode. Each adapter c an control up to three SCS I channels. Internal and exter nal disk drives ca n be mixed on channels 0 and 1, but not on channel 2.

SCSI Drive St at e A SCSI physical drive can be i n one of these states:

· Online: Powered-on and operational.

· Ready: Functioning no rmally, but not par t of a configured logic al

drive and not designated as a hot spare.

· Hot Spare: Powered-on stand-by disk drive, ready for use if an

online disk fails.

· Fail: Out of ser vice, due to a fault occurring on the drive.

· Rebuild: Currently being rebuilt with data from a failed drive.

B-7

Service Provider

The Service Provider (SP) is a program that resides in the desktop system or ser ver and is responsible for all DMI a ctivities. This layer collects manage ment information from prod ucts (whether system hardware, peripherals, or software), stores that information in th e DMI database, and passes it to management applications as requested.

SNMP Simp le Network Management Protocol. The most wide ly used protocol

for communicating manageme nt information between the man aged elements of a network and a networ k man ager. SNMP focuses primar ily on the network backbone. The Inter net standard protocol that manages nodes on an Inter net Protocol (IP) net work.

Spanning Array spann ing by a logical dr ive combines s torage space in two a rrays

of disk drives into a si ngle, contiguous storage spac e in a logical dr ive. MegaRAID logical d rives can span consecuti vely numbered arrays that each consist of the sam e number of disk drives. Array spanning promotes RAID levels 1 and 5 to RAID levels 10 and 50, respectively. See also Disk Spann ing.

Spare A disk drive available to back up the data of other dr ives. Stripe Size The amount of data contigu ously written to eac h disk. You c an specify

stripe sizes of 4, 8, 16, 32, 64, and 128 K bytes for each logical drive. For best performance, choose a stripe size equal to or smaller than the block size used by the host computer.

Stripe Width The number of disk drives across which the d ata is striped. Striping Segmentat ion of logically seque ntial data, such as a single file, so that

segments can be wr itten to multiple physical devices in a round-robin fashion. This technique is use ful if the processor can read or write data faster than a single disk c an supply or accept it . While data is being transferred from the first disk, the s econd disk can locate the next segment. Data striping is used in some modern databases and in certain RAID devices.

Terminator A resistor connected to a signal wire in a bus or network for impedance

matching to prevent reflections—for example, a 50 ohm res istor connected across t he end of an Ethern et cable. SCSI buses and some LocalTalk wiring schemes also require terminators.

B-8 Glossary

Ultra320 A sub set of Ultra3 SCSI that allows a ma ximum throughput of 320

Mbytes/s, which is twice as fast as Wide Ultra2 SCSI. Ultra320 SCSI provides 320 Mbytes/s on a 1 6-bit connection.

Virtual Sizing FlexRAID vir tual siz ing is use d to crea te a logic al dri ve up to 80 Gbytes.

A maximum of 40 logical dr ives can be confi gured on a RAID controller, and RAID migration is poss ible for all logical drives except the for tieth. Because it is no t possible to do migration on the las t logical drive, the maximum space available for RAID migration is 560 Gbytes.

Wide SCSI A variant on the SCSI-2 inter face. Wide SCSI uses a 16-bit bus, double

the width of the or iginal SCSI-1. Wid e SCSI devices cannot be connected to a SCSI-1 bus. Like F ast SCSI, Wide SCSI supports transfer rates up to 20 Mbytes/s.

B-9

B-10 Glossary

Index

A

Array B-1 Array Configurati on Planner Array management sof tware Array performance fea tures Array spanning Assigning RAID levels Asynchronous operations Audible Warnings Automatic failed dri ve detection and re build

B-1

A-1

5-11

B-1

5-14

B-1

4-3

B

BIOS 4-6 BIOS Boot Error Mess ages BIOS Configuration Utility BIOS message Bus data transfer rate Bus type Bus-based

6-6

4-6

2-11

7-3

6-7

4-6

C

Cache configuration 4-6 Cache I/O Cache Memory Card size Channel Cold swap Compatibility Configuration on Disk Configuration Strate gies Configuring Arrays Configuring Logical Drives Configuring SCSI phys ical drives Consistency check CPU Creating hot spares Creating logical drives

4-6

B-1

4-6

B-2

4-9

4-7

5-8 2-4, B-2

5-8

4-2

5-9

5-12

5-1

D

Data redundancy

Using mirroring Data transfer capacity Dedicated parity Degraded Disconnect/reconnect Disk

B-2

Disk array

2-8

2-10, B-2

B-2

2-6

B-2

4-7

4-4

Disk Array Types Disk duplexing Disk mirroring Disk rebuild Disk spanning Disk striping Disk subsystem Documentation DOS

4-5

Double buffering Drive roaming Drivers

6-7

2-11

B-2

2-6, B-2

2-9

2-7, B-2

2-4, B-2

B-3

1-4

B-3

4-2

E

Enclosure manageme nt 2-11

F

Fail 2-10

2-10

Failed Failed drive Fast SCSI Fault tolerance Fault tolerance feat ures Features Firmware Flash ROM FlexRAID Power Fail option Format

B-3

1-2, 4-1

4-6, B-3

1-2

B-3

2-4

4-4

G

GB B-3

B-1

Glossary

H

Hardware Installation 6-1 Host computer Host-based arra y Host-based RAID solution Hot spare

Using during disk rebuild

Hot swap

B-4

B-3

2-8, 2-10, B-4

2-9, 4-4, B-4

2-2

I

I/O driver B-4 Initialization

B-4

B-3

2-9

MegaRAID SCSI 320-0 Hardware Guide IX-1

Install Drivers 6-7 Installation steps

Custom

6-2

J

Jumpers

Setting

6-3

L

Linux

Red Hat

4-5

Logical disk Logical drive Logical drive states

B-4

2-3, B-4

2-10

M

Mapping B-4

B-5

MB MegaRAID BIOS MegaRAID BIOS Configuration Utility MegaRAID Manag er MegaRAID SCSI 320 -0 card

Installing MegaRAID SCSI 320 -0 card layout Mirroring MS-DOS Multi-threaded

4-7

4-8

6-5

2-6 6-7

B-5

N

Nonvolatile RAM 4-6 Novell NetWare NVRAM

1-2

4-5, 6-7

O

Offline 2-10 Online

Drive state Operating environment Operating system software d rivers Operating voltage Optimal Optimizing Data Stor age Other BIOS error me ssages

2-10

4-6

B-5

5-12

4-5

7-4

P

Parity 2-8, B-5

B-5

Partition PCI controller Physical array Physical disk Physical disk roaming Physical drive Power Console Plus Processor Product specifications Protocol

B-5

4-6

2-3

B-5

2-3

B-5

4-8

4-6

4-8

6-3

R

RAID B-6

RAID 0 RAID 1

RAID 10 RAID 5 RAID 50 RAID benefits

RAID levels RAID levels supported RAID management RAID management feature s RAID migration RAID overview Read-ahead Ready Ready state Rebuild Rebuild rate Rebuilding a disk Reconnect Reconstruct Reconstruction Red Hat Linux RedAlert Redundancy Replacement disk Replacement unit

2-1

Benefits Introduction to

2-1 3-2 3-3

Spanning to configu re RAID 10

3-6

Configuring Spanning to make RAID 50 Configuring Improved I/O

Increased reliability

2-8

3-4

3-7

2-8

1-v, 3-1, B-6

2-8

2-1

2-2

4-6

4-8

4-4

B-6

2-3

B-6

2-10

B-6

2-10, 4-7

2-9, B-6

2-9

4-7

B-6

6-7

4-9

B-7

2-8

S

SAF-TE B-7

B-7

SCSI SCSI backup and utility software SCSI bus widths and maximum throughp ut SCSI buses SCSI channel SCSI data transfer rate SCSI Device Compatibility SCSI device types supported SCSI Drive State SCSI firmware SCSI to SCSI SCSI-to-SCSI RAID product Server managemen t Service provider SMART Technology SNMP SNMP agent SNMP managers Software utilities Software-based Spanning Spare Standby rebuild Stripe size Stripe width

1-2

B-7

2-11

B-8

4-9

2-7, B-8

B-8

2-6, 4-7, B-8

2-5, B-8

4-7

B-8

4-5

2-11

2-9

B-7

4-9

4-1

4-9

4-6

4-9

4-6

2-3

1-4

IX-2 Index

B-8

Striping

T

Technical Support 1-vi Terminator Troubleshooting

B-8

7-1

U

Ultra320 and Wide SCSI 4-1 Ultra3-SCSI (320M) Unpack

6-2

B-9

V

Virtual sizing B-9

W

WebBIOS Configuration Utility 4-8 Wide SCSI Windows .NET Windows 2000 Windows NT Windows XP

B-9

6-7

6-7 4-5, 6-7 6-7

Index IX-3

IX-4 Index

Customer Feedback

We would appreciate your feedback on this document. Please copy the following page, add your comments, and fax it to us at the number shown.

If appropriate, please also fax copies of any mar ked-up pages from this document.

Important:

Thank you for your help in improving the q uality of our documents.

Please include your name, phone number, fax number, and company address s o that we may contact you directly for clarification or additional informa tion.

MegaRAID SCSI 320-0 Hardware Guide 1-1

Reader’s Comments

Fax your comments to: LSI Logic Cor poration

Te chnical Publications M/S E-198 Fax: 408.433.4333

Please tell us how you rate this document: MegaRA ID SCSI 320- 0 Hard- ware Guid e. Place a check mar k in the appropriate blank for each category.

Excellent Good Average Fair Poor

Completeness of information ____ ____ ____ ____ ____ Clarity of information ____ ____ ____ ____ ____ Ease of finding information ____ ____ ____ ____ ____ Technical content ____ ____ ____ ____ ____ Usefulness of examples and

illustrations

Overall manual ____ ____ ____ ____ ____

____ ____ ____ ____ ____

What could we do to imp rove this document?

If you found errors in this documen t, please specify the err or and page number. If appropriate, please fax a mar ked-up copy of the page(s).

Please complete the i nformation below so that we may contact you directly for clarific ation or additional inform ation.

Name Date Telephone Title Department Mail Stop Company Name Street City, State, Zip

1-2 Customer Feedback

Fax

Acer SCSI 320-0 User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Preface

Contents

1.1 Features

1.2 NVRAM and Flash ROM

1.3 Single-Ended and Differential SCSI Buses

1.4 Maximum Cable Length for SCSI Standards

1.5 SCSI Bu s Widths and Maximum Throughput

1.6 Documentation

1.6.1 MegaRAID SCSI 320-0 ZCR Hardware Guide

1.6.2 MegaRAID Configuration Software Guide

1.6.3 MegaRAID Operating System Driver Installation Guide

2.1 RAID Benefits

2.1.1 Improved I/O

2.1.2 Increased Reliability

2.2 MegaRAID SCSI 320-0 – Host-Based RAID Solution

2.2.1 Host-Based RAID

2.2.2 SCSI-to-SCSI External RAID

2.3 RAID Overview

2.3.1 Physical Array

2.3.2 Logical Drive

2.3.3 Consistency Check

2.3.4 Fault Tolerance

2.3.5 Disk Striping

2.3.5.1 Stripe Width

2.3.5.2 Stripe Size

2.3.6 Disk Mirroring

2.3.7 Disk Spanning

2.3.8 Parity

2.3.9 Hot Spares

2.3.10 Hot Swapping

2.3.11 Disk Rebuild

2.3.11.1 Rebuild Rate

2.3.12 Logical Drive States

2.3.13 SCSI Drive States

2.3.14 Disk Array Types

2.3.15 Enclosure Management

3.1 Selecting a RAID Level

3.2 RAID 0

3.3 RAID 1

3.4 RAID 5

3.5 RAID 10

3.6 RAID 50

4.1 SMART Technology

4.2 Configur ation on Disk

4.3 Configur ation Features

4.4 Array Performance Features

4.5 RAID Management Features

4.6 Fault Tolerance Features

4.7 Software Utilities

4.8 Operating System Software Drivers

4.9 MegaRAID SCSI 320-0 Specifications

4.9.1 PCI Bridge/ CPU

4.9.2 Cache Memory

4.9.3 MegaRAID BIOS

4.9.4 SCSI Firmw are

4.10 RAID Management

4.10.1 MegaRAID BIOS Configuration Utility

4.10.2 WebBIOS Configuration Utility

4.10.3 Power Console Plus

4.10.4 MegaRAID Manager

4.11 Compatibility

4.11.1 Server Management

4.11.2 SCSI Device Compatibility

4.11.3 Software

5.1 Confi guring SCSI Physical Drives

5.1.1 Basic Configuration Rules

5.1.2 Current Physical Device Configuration

5.1.3 Logical Drive Configuration

5.1.4 Physical Device Layout

5.2 Configur ing Arrays

5.2.1 Arranging Arrays

5.2.2 Creating Hot Spares

5.3 Creating Logical Drives

5.3.1 Configuration Strategies

5.3.1.1 Maximize Capacity

5.3.1.2 Maximizing Drive Availability