HP Smart Array Controller technology

HP Smart Array Controller technology
technology brief, 3rd edition
Abstract.............................................................................................................................................. 3
Introduction......................................................................................................................................... 3
Storage trends..................................................................................................................................... 3
Meeting data storage requirements........................................................................................................ 4
High performance................................................................................................................................ 4
PCI Express technology..................................................................................................................... 4
SAS/SATA technology ..................................................................................................................... 5
SAS-2 standard ............................................................................................................................ 6
Mini SAS 4x cable connectors and receptacles................................................................................ 7
High-performance processor.............................................................................................................. 7
Cache module benefits ..................................................................................................................... 8
Read-ahead caching..................................................................................................................... 8
Write-back caching ...................................................................................................................... 8
Balanced cache size..................................................................................................................... 9
RAID performance enhancements....................................................................................................... 9
Disk striping................................................................................................................................. 9
Parity data................................................................................................................................... 9
Background RAID creation............................................................................................................. 9
RAID 5 and RAID 6 read-modify-write ............................................................................................. 9
Striping across arrays ................................................................................................................. 10
RAID 1 load balancing ............................................................................................................... 10
Hardware versus software RAID ................................................................................................... 10
Smart Array performance................................................................................................................ 11
Data availability ................................................................................................................................ 14
RAID support ................................................................................................................................. 14
Stripe size migration ................................................................................................................... 14
RAID migration .......................................................................................................................... 15
Drive roaming ............................................................................................................................... 16
Mirror splitting and recombining...................................................................................................... 16
Online drive flash .......................................................................................................................... 16
Recovery ROM .............................................................................................................................. 16
Pre-Failure Warranty using S.M.A.R.T technology .............................................................................. 16
Automatic data recovery with rapid rebuild technology ...................................................................... 17
Online spare ................................................................................................................................. 17
Dynamic sector repair .................................................................................................................... 18
ECC protection .............................................................................................................................. 18
Battery-backed write cache ............................................................................................................. 18
Recovering data from battery-backed cache .................................................................................. 19
Selection criteria for battery-backed cache .................................................................................... 19
Types of batteries ....................................................................................................................... 19
Battery replacement .................................................................................................................... 20
Alternatives to battery replacement ............................................................................................... 20
Flash-backed write cache ................................................................................................................ 20
FBWC architecture ..................................................................................................................... 21
FBWC cache ............................................................................................................................. 22
Super-capacitor.......................................................................................................................... 22
Capturing data during power loss ................................................................................................ 22
Recovering data from the flash-backed cache ................................................................................ 23
Entry level RAID solutions ................................................................................................................ 23
Software RAID ........................................................................................................................... 23
Zero Memory RAID..................................................................................................................... 23
Storage support and pathway redundancy ........................................................................................... 24
Solid state drives............................................................................................................................ 24
Native Command Queuing ............................................................................................................. 24
Dual Domain support...................................................................................................................... 24
Tape device support ....................................................................................................................... 25
Smart Array Advanced Pack ............................................................................................................... 25
Storage management ......................................................................................................................... 26
Array Configuration Utility .............................................................................................................. 27
Option ROM Configuration for Arrays.............................................................................................. 27
CPQONLIN................................................................................................................................... 27
HP Systems Insight Manager ........................................................................................................... 27
Performance monitoring .............................................................................................................. 28
Fault prediction .......................................................................................................................... 28
Array Diagnostics Utility.................................................................................................................. 28
Summary .......................................................................................................................................... 28
Appendix A: Capacity growth technologies .......................................................................................... 29
Array expansion ............................................................................................................................ 29
Logical drive creation ..................................................................................................................... 30
Logical drive extension ................................................................................................................... 31
For more information.......................................................................................................................... 32
Call to action .................................................................................................................................... 32
Abstract
This technology brief describes specific functions of the HP Smart Array controller family and explains
how Smart Array technology meets administrators’ requirements for capacity growth, high
performance, data availability, and manageability.
Introduction
In today’s networking environments, administrators face difficult online data storage problems and
ever-increasing performance demands. While PCI Express (PCIe) provides a high speed interface for
data communication, the HP family of parallel SCSI, Serial Attached SCSI (SAS), and Serial ATA
(SATA) Smart Array controllers addresses direct attach and storage area network (SAN) requirements.
References made in this technology brief to the ‘present generation’ of Smart Array controllers refer to
the P410i, P410, P411, P212, and P712. These SAS-based, PCIe 2.0 compliant Smart Array
controllers were released in the first half of 2009.
HP Smart Array controllers provide administrators with configuration, management, and diagnostics
tools. These tools provide high levels of usability as well as consistency between generations of
products. This continuity ensures that administrators can move data between servers and external
storage enclosures, and between models of Smart Array controllers.
For complete Smart Array controller compatibility and support information, see
www.hp.com/products/smartarray
Storage trends
Four key trends influence network storage requirements today:
• Application complexity — Network applications are becoming more complex, leading to larger
files and the need to maintain more information online for immediate user access.
• Mission-critical migration — More mission-critical data is moving to servers, either down from larger
systems or up from paper-based processes.
• Server consolidation — Multiple servers and applications are being consolidated onto fewer servers
for increased control and centralized network management.
• Efficiency expectations — Corporate information technology organizations are managing more
applications and more data while at the same time reducing the size of their administrative staffs.
These key trends generate five primary data storage requirements:
• Capacity growth — Storage solutions must provide not only adequate capacity for today’s
applications but also flexibility for future growth.
• High performance — Data storage subsystems must deliver enough performance to accommodate
an increasing number of users while maintaining rapid response times. In many environments, the
storage subsystem is the most critical determinant of overall system performance.
• Data availability — Because businesses depend on their mission-critical data, it must be accessible
at all times to maintain user productivity.
• Manageability — Storage solutions must reduce the total cost of ownership by making network
storage management intuitive, informative, and less time consuming.
• Virtualization — Virtualization improves storage efficiency and simplifies operation by pooling,
sharing, and centrally managing SAN storage.
3
Meeting data storage requirements
The HP Smart Array controller family has an advanced intelligent architecture developed with a
feature set that specifically addresses today’s network data storage requirements:
• Capacity growth — Administrators can expand server storage capacity while the server is
operating. The HP online Array Configuration Utility (ACU) simplifies storage management for
administrators who use fault-tolerant array configurations and experience capacity growth. Refer to
Appendix A for details about using array creation and expansion to increase capacity.
• High performance — Smart Array controllers supply high levels of throughput and data bandwidth
to support demanding workloads. The advanced intelligent architecture contains specialized
modules that enhance performance significantly. Online array expansion, stripe size migration,
RAID migration, configurable read/write cache sizes, rebuild priority, and configurable write-back
caching for each logical drive empowers administrators to tune Smart Array controllers to meet
performance demands.
• Data availability — Smart Array controllers include a comprehensive set of fault-tolerance features
that guard against disk drive and controller component failures, making online storage
administration possible and reducing planned downtime.
• Manageability — All Smart Array controllers use a common set of management software, reducing
training requirements and technical expertise necessary to install and maintain HP server storage.
The common data format among Smart Array controllers means customers can easily upgrade to
future Smart Array products.
• Storage virtualization — HP StorageWorks SAN Virtualization Services Platform (SVSP) provides
centrally managed storage pooling and virtual volume provisioning for HP and non-HP storage
resources. The HP SVSP enables high scalability, top performance, and high availability of all SAN
resources. It brings substantial improvements to managing enterprise storage, reducing
administration costs, and improving asset utilization. It also offers a consistent set of storage and
mobility services that improves flexibility, protects data, and decreases business risk. Services
offered include volume management, data migration, copy services (clones and snapshots),
synchronous and asynchronous mirroring, and thin provisioning. More information on HP storage
virtualization products and services are available at
http://h18006.www1.hp.com/products/storage/software/sanvr/index.html
High performance
Smart Array controllers offer exceptional performance and reliability characteristics, with support for
traditional parallel Small Computer System Interface (SCSI) technology, as well as the latest SAS and
SATA technology and advanced RAID capabilities.
PCI Express technology
The present generation of Serial Attached SCSI (SAS) based Smart Array controllers released in the
first half of 2009, and all ProLiant G6 servers support the PCIe 2.0 specification. PCIe 2.0 has a perlane signaling rate of 5 Gb/s ― double the per-lane signaling rate of PCIe 1.0 (Figure 1).
4
Figure 1. PCIe data transfer rates
Lane 1 Send
Lane 1 Receive
Source
Target
Lane n Send
Lane n Receive
Max. bandwidth
(Send or receive)
Total
(Send and receive)
Link
size
PCIe 1.0
PCIe 2.0
PCIe 1.0
PCIe 2.0
x1
250 MB/s
500 MB/s
500 MB/s
1 GB/s
x4
1 GB/s
2 GB/s
2 GB/s
4 GB/s
x8
2 GB/s
4 GB/s
4 GB/s
8 GB/s
x16
4 GB/s
8 GB/s
8 GB/s
16 GB/s
NOTE:
10 bits of signaling are required to transport one Byte of data. Consequently in Figure
1, 5 Gb/s is shown as 500MB/s.
PCIe 2.0 is completely backward compatible with PCIe 1.0. A PCIe 2.0 device can be used in a PCIe
1.0 slot and a PCIe 1.0 device can be used in a PCIe 2.0 slot. Table 2 shows the level of
interoperability between PCIe cards and PCIe slots.
Table 1. PCIe device interoperability
PCIe
device type
x4 Connector
x4 Link
x8 Connector
x4 Link
x8 Connector
x8 Link
x16 Connector
x8 Link
x16 Connector
x16 Link
x4 card
x4 operation
x4 operation
x4 operation
x4 operation
x4 operation
x8 card
Not allowed
x4 operation
x8 operation
x8 operation
x8 operation
x16 card
Not allowed
Not allowed
Not allowed
x8 operation
x16 operation
SAS/SATA technology
The newest serial, PCIe 2.0 capable Smart Array controllers use Serial Attached SCSI (SAS)
technology, a point-to-point architecture in which each device connects directly to a SAS port rather
than sharing a common bus as with parallel SCSI devices. Point-to-point links increase data
throughput and improve the ability to locate and fix disk failures. More importantly, SAS architecture
solves the parallel SCSI problems of clock skew and signal degradation at higher signaling rates. 1
The same Smart Array controllers are compatible with Serial Advanced Technology Attachment
(SATA) technology and include the following features to enhance performance and maintain data
availability and reliability:
• SAS and SATA compatibility — The ability to use either SAS or SATA disk drives lets administrators
deploy drive technology that fits each computing environment. HP Smart Array controllers can
manage both SAS arrays and SATA arrays. Smart Array configuration utilities help administrators
configure arrays correctly so that data remains available and reliable.
For more information about SAS technology, refer to the HP technology brief titled “Serial Attached SCSI storage technology”
available at http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01613420/c01613420.pdf .
1
5
• SAS wide port operations — Wide ports contain four single lane (1x) SAS connectors and the
cabling bundles all four lanes together. SAS wide ports allow balanced SAS traffic distribution
across the links for enhanced performance. In addition, wide ports provide redundancy by
tolerating up to three physical link failures while maintaining the ability to communicate with the
disk drives. The most common use of wide links is to a JBOD or to an internal server expander
connecting to large numbers of drives. No special configuration is required for this functionality.
• SAS expanders — SAS expander devices offer higher system performance by expanding the
number of disk drives that can be attached to an HP Smart Array controller. SAS expanders are an
aggregation point for large numbers of drives or servers providing a common connection. By
cascading expanders, administrators can chain multiple storage boxes together.
The full height, HP Serial Attached SCSI (SAS) expander card supports up to 24 internal drive bays
plus an external connection for tape. The HP SAS Expander Card is well suited for ProLiant server
customers that want to use RAID with more than eight internal disk drives. With the SAS expander
installed between the drives and the controller, users will see the same performance characteristics as
if the drives were connected to JBOD. For more information on the HP SAS Expander Card, go to
http://h18004.www1.hp.com/products/servers/proliantstorage/arraycontrollers/sas-expander/index.html.
SAS-2 standard
The second-generation SAS (SAS-2) link speed 2 of 6 Gb/s is double the SAS-1 transfer rate. SAS-2
link speeds require SAS-2 compliant disk drives. SAS-2 eliminates the distinction between fanout and
edge expanders by replacing them with self-configuring expanders. SAS-2 enables zoning for
enhanced resource deployment, flexibility, security, and data traffic management.
SAS-2 connections have the potential to deliver peak raw data bandwidth of up to 600 megabytes
per second (MB/s) per physical link in each direction. SAS-2 devices are capable of sending and
receiving data simultaneously across each physical link, which is known as full duplex. When
effectively implemented, full duplex, 6 Gb/s SAS connections can deliver peak raw data bandwidth
of up to 1200 MB/s between the controller and storage device. It is important to note that the SAS-2
data bandwidths described here are theoretical speeds identified by the SAS-2 standard. Real world
performance will be affected by the performance of storage devices attached to the SAS-2
connection.
Smart Array controllers, with releases beginning in the first quarter of 2009, incorporate SAS-2
capable performance and features. The SAS-2 standard maintains compatibility with both Serial SCSI
and Serial ATA protocols for communicating commands to SAS and SATA devices. SAS-2 compliant
controllers are compatible with present generation 3Gb/s SATA & 6Gb/s SAS as well has
backwards compatible with past generation SATA & SAS devices.
The Smart Array P700m controller supports SAS-2 zoning capabilities. SAS-2 zoning uses a shared
SAS fabric to assign multiple servers to multiple storage devices within a rack. HP presently employs
SAS-2 zoning in the HP 3Gb SAS BL Switch & MDS600 drive enclosure solution with the Smart Array
P700m. This allows the creation of a group of up to 100 drive bays across four MDS600 drive
enclosures. The group of drive bays can then be electronically assigned to any server within the cClass enclosure. 3
For an up-to-date listing of HP Smart Array controllers that support the SAS-2 specification, see the
Smart Array controller matrix: www.hp.com/products/smartarray
Serial Attached SCSI-2 (SAS-2) is an American National Standards Institute (ANSI) standard from the INCITS T10 Technical
Committee on SCSI Storage Interfaces. SAS-2 is the successor to SAS-1.1 and SAS-1.
3
For more information on BladeSystem Direct Connect SAS configurations, go to
http://www.hp.com/hpinfo/newsroom/press_kits/2009/convergeeverything2009/DataSheetDirectconnectSASstorageforHPB
ladeSystem.pdf
2
6
Mini SAS 4x cable connectors and receptacles
Mini SAS 4x connectors and receptacles (Figure 2) are replacing SAS 4x connectors and receptacles
in present generation Smart Array controllers. The ground pins in Mini SAS connectors can be used
for power in active cables.
Figure 2. The Mini SAS 4i connector (left) and receptacle (right) are replacing SAS 4x connectors and
receptacles.
For more detailed information on SAS technology and SAS-2 zoning, refer to the Serial Attached
SCSI storage technology brief:
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01613420/c01613420.pdf
High-performance processor
HP Smart Array controllers use a variety of high-performance processors for managing the RAID
system. The Power PC and MIPS 4 processors are the most widely used among Smart Array
controllers. The PowerPC processor is a 64-bit processor based on reduced instruction set computer
(RISC) technology. The processor connects to its internal peripherals using separate 128-bit-wide readand-write buses, each running at 133 MHz.
Power PC processors have a highly pipelined architecture that allows dual instruction fetch, decode,
and out-of-order issue, as well as out-of-order dispatch, execution, and completion. The Power PC
flexibility features include independently configurable data cache arrays, write-back and write-through
operation, and performance characteristics that increase cache memory allocation efficiency.
Designed for extensive power management and maximum performance, Power PC processors
provide increased throughput and a streamlined path to instruction completion, which means faster
user access to data.
The present generation Smart Array P410, P411, and P212 controllers use an embedded PM8011
SRC 8x6G 6Gb/s SAS RAID-on-Chip (RoC) MIPS processor. This reduced instruction set computing
(RISC), instruction set architecture (ISA) processor operates at 600 MHz, has 34K instruction and data
caches, and a 32-bit multi-threading I/O. The MIPS processor uses a 4-way set associative writeback, and flexible thread policy manager with programmable quality of service (QoS) support.
4
Microprocessor without Interlocked Pipeline Stages
7
Cache module benefits
With advanced read-ahead and write-back caching capabilities, the Smart Array controller cache
module produces significant performance improvements for I/O operations.
Read-ahead caching
The HP Smart Array controller family uses an intelligent read-ahead algorithm that can anticipate data
needs and reduce wait time. It can detect sequential read activity on single or multiple I/O threads
and predict when sequential read requests will follow. The algorithm then “reads ahead,” or prefetches data, from the disk drives before the data is actually requested. When the read request
occurs, the controller retrieves the data from high-speed cache memory in microseconds rather than
from the disk drive in milliseconds.
This adaptive read-ahead scheme provides excellent performance for sequential small block read
requests. At the same time, the system is not penalized by random read patterns because read-ahead
functionality is disabled when non-sequential read activity is detected. Thus, Smart Array controller
technology overcomes the problem with some array controllers in the industry today that use fixed
read-ahead schemes to increase sequential read performance but degrade random read
performance.
Write-back caching
HP Smart Array controllers also use a caching scheme that allows host applications to continue
without waiting for write operations to complete to the disk. This technique is also called postedwrites, or write-back caching. Without this type of caching, the controller must wait until write data is
actually written to disk before returning completion status to the operating system. With write-back
caching, the controller can “post” write data to high-speed cache memory and immediately return
“back” completion status to the operating system. The write operation is completed in microseconds
rather than milliseconds.
Once write data is located in the cache, subsequent reads to the same disk location will be sourced
from the cache. Subsequent writes to the same disk location will replace the data held in cache. This
technique is called a “read cache hit.” This feature improves bandwidth and latency for applications
that frequently write and read the same area of the disk.
After write data is located in the cache, the controller finds opportunities to combine adjacent write
transfers into a larger transfer that can efficiently move to the disk drive. This technique is called “write
coalescing.” Disk drives achieve higher throughput when processing many small transfers rather than
fewer large transfers. This feature improves write bandwidth whenever data is sequentially written to
the controller using write command sizes smaller than the stripe size of the logical drive.
With write data located in the cache, logical drives in RAID 5 and RAID 6 configurations achieve
higher write performance by combining adjacent write requests to form a full stripe of data. This
technique is called “full-stripe writes.” Write operation for RAID 5 and RAID 6 normally requires extra
disk reads to compute the data for the parity drives. However, if all the data required for a full stripe
is available in the cache, the controller does not require the extra disk reads. This feature improves
write bandwidth whenever data is sequentially written to a logical drive in a RAID 5 or RAID 6
configuration.
Data in the controller’s write cache is written to disk later, at an optimal time for the controller. While
the data is in cache, it is protected against memory chip failure by error checking and correction
(ECC) DRAM technology, and against system power loss by the integrated battery backup
mechanism. Smart Array controllers avoid the risk of data loss by ensuring that the battery backup is
present before enabling write-back cache. Disk drives provide an option to enable write-caching that
is not battery backed. HP advises against enabling disk drive write cache because a power or
equipment outage could result in data loss.
8
Balanced cache size
Smart Array controllers allow administrators to adjust how the cache is distributed for write-back and
read-ahead operations. Administrators can configure the cache module for optimal performance for
any storage need. The default setting for most present generation Smart Array controllers configures
the cache for 75 percent write-back operations and 25 percent read-ahead operations, but these
default settings can vary by controller. Additionally, the cache module capacity can be upgraded to
increase caching performance.
RAID performance enhancements
Smart Array controllers use several enhancements to increase RAID performance.
Disk striping
Striping combines several individual disk drives into a larger disk array containing one or more
logical drives. Performance of the individual drives is aggregated to provide a single highperformance logical drive. The array controller evenly distributes the logical drive data into small
“stripes” of data sequentially located across each member disk drive in the array. Administrators can
adjust the stripe size to achieve optimal performance. Performance improves as the number of drives
in the array increases.
Parity data
In a RAID 5 configuration, data protection is provided by distributed parity data. This parity data is
calculated stripe by stripe from the user data that is written to all other blocks within that stripe. The
blocks of parity data are distributed evenly over every physical drive within the logical drive. When a
physical drive fails, data that was on the failed drive can be calculated from the remaining parity
data and user data on the other drives in the array. This recovered data is usually written to an online
spare drive through a process called a rebuild.
RAID 6, like RAID 5, generates and stores parity information to protect against data loss caused by
drive failure. With RAID 6, however, two different sets of parity data are used so that data can still be
preserved even if two drives fail. Each set of parity data uses a capacity equivalent to that of one of
the constituent drives. This method is most useful when data loss is unacceptable but cost is also an
important factor. RAID 6 provides better protection for data than a RAID 5 configuration because of
the additional parity information.
Background RAID creation
When a RAID 1, RAID 5, or RAID 6 logical drive is first created, the Smart Array controller must build
the logical drive within the array before enabling certain advanced performance techniques. While
the logical drive is created, the storage volume is accessible by the host with full fault tolerance. The
Smart Array controller creates the logical drive whenever the controller is not busy; this is called
background parity initialization. Parity initialization takes several hours to complete, depending on
the size of the logical drive and how busy the host keeps the controller. Before parity initialization
completes, normal writes to RAID 5 and RAID 6 logical drives are slower because the controller must
read the entire stripe to update the parity data and maintain fault tolerance. These writes during
parity initialization are called regenerative writes or reconstructed writes.
RAID 5 and RAID 6 read-modify-write
After parity initialization is complete, writes to a RAID 5 or RAID 6 logical drive are typically faster
because the controller does not read the entire stripe to update the parity data. Since the controller
knows that the parity data is consistent with all the member drives in the stripe, the controller needs to
read from only two disk drives during a RAID 5 write (or three disk drives for a RAID 6 write) to
compute the parity data (regardless of array size). This technique is called a read-modify-write or
backed-out write.
9
Striping across arrays
RAID 50 and 60 methods stripe the data across multiple RAID/JBOD sets with different levels of
parity. These nested RAID types allow users to configure arrays across HP Modular Smart Arrays
(MSAs).
RAID 50 (RAID 5+0) is a nested RAID method that uses RAID 0 block-level striping across RAID 5
arrays with distributed parity. RAID 50 will tolerate one drive failure in each spanned array without
loss of data. RAID 50 configurations require a minimum of six drives and require less rebuild time
than single RAID 5 arrays.
RAID 60 (RAID 6+0) is a nested RAID method that uses RAID 0 block-level striping across multiple
RAID 6 arrays with dual distributed parity. With the inclusion of dual parity, RAID 60 will tolerate the
failure of two disks in each spanned array without data loss. RAID 60 configurations require a
minimum of eight drives.
RAID 6 and 60 are available as an option with the Smart Array Advanced Pack (see section later in
this paper) and are not supported on all HP Smart Array controllers.
RAID 1 load balancing
In general, the same stripe and array sizes, RAID 0, RAID 5, and RAID 6, have the same read
performance. RAID 1 logical drives contain two copies of the data. During reads to RAID 1 logical
drives, the Smart Array controller issues read requests to either drive in the mirrored set. During a
heavy read load, the Smart Array controller balances the number of requests between the two disk
drives to achieve higher read bandwidth. This technique is called RAID 1 load balancing.
Hardware versus software RAID
Today’s operating systems offer basic support for RAID 0, RAID 1, and RAID 5 disks (called software
RAID) to create and manage logical drives that do not contain the operating system. Software RAID
requires a significant amount of the server’s resources to perform management functions. However,
Smart Array controllers use a separate processor and memory subsystem for management functions.
Furthermore, with Smart Array controllers, the parity calculations required by RAID 5 and RAID 6 are
performed by specialized hardware engines that maximize data throughput for disk write, rebuild,
and regenerate read operations.
10
Smart Array performance
This section presents results of HP testing for sequential read/write performance between Smart Array
controller P411, announced in April 2009, and the previous generation P800. The charts in Figures
3, 4, and 5 compare the maximum performance for each Smart Array controller across the queue
depth in each data range. The same queue depth (1, 2, 4, 8, 16, 32, 64, 128, 256) and stripe size
(256 KB) was used for all testing.
It should be noted that the two controllers operated under different PCI bandwidth constraints. While
the DL380 G5 server/P800 configuration operated at 3Gb SAS PCIe Gen 1 speeds, the DL380 G6
server/P411 configuration operated at 6Gb SAS PCIe-2 speeds.
The maximum sequential write performance of the P411in the RAID 0 test, Figure 3, was
approximately 100 percent higher than that of the P800. These performance gains are largely due to
bandwidth differences between 3Gb SAS PCI Gen 1 and 6Gb SAS PCIe-2.
Figure 3. Maximum sequential read/write performance for the Smart Array P800 and P411
controllers in a RAID 0 environment
RAID-0, Maximum performance
2500
DATA RATE (MB Per Second)
2000
1500
1000
500
0
4KB
SEQ
READ
64KB
SEQ
READ
512KB
SEQ
READ
1MB
SEQ
READ
4KB
SEQ
WRITE
64KB
SEQ
WRITE
128KB
SEQ
WRITE
256KB
SEQ
WRITE
512KB
SEQ
WRITE
P411, 512MB, fw:2.50, RAID 0 (256KB), D2700, 14-146GB SFF 6Gb SAS 15K, DL380-G6
P800, 512MB, fw:7.00, RAID 0 (256KB), MSA70, 14-146GB SFF 3Gb SAS 15K, DL380-G5
11
The performance gains shown by the Smart Array P411 in RAID 5 and 6 sequential writes (Figures 4
and 5) result from improvements to the Smart Array firmware and hardware. The RAID 5 maximum
sequential read/write performance test results shown in Figure 4 indicate significant performance
gains for the Smart Array P411 over the previous generation P800 controller.
Figure 4. Maximum sequential read/write performance for the Smart Array P800 and P411 controllers
in a RAID 5 environment
RAID-5, Maximum performance
2500
DATA RATE (MB Per Second)
2000
1500
1000
500
0
4KB
SEQ
READ
64KB
SEQ
READ
512KB
SEQ
READ
1MB
SEQ
READ
4KB
SEQ
WRITE
64KB
SEQ
WRITE
128KB
SEQ
WRITE
256KB
SEQ
WRITE
512KB
SEQ
WRITE
P411, 512MB, fw:2.50, RAID 5 (256KB), D2700, 14-146GB SFF 6Gb SAS 15K, DL380-G6
P800, 512MB, fw:7.00, RAID 5 (256KB), MSA70, 14-146GB SFF 3Gb SAS 15K, DL380-G5
Again, the RAID 6 maximum sequential read/write performance test results shown in Figure 5
indicate significant performance gains for the Smart Array P411 over the previous-generation P800
controller.
12
Figure 5. Maximum sequential read/write performance for the Smart Array P800 and P411 controllers
in a RAID 6 environment
RAID-6, Maximum performance
2500
DATA RATE (MB Per Second)
2000
1500
1000
500
0
4KB
SEQ
READ
64KB
SEQ
READ
512KB
SEQ
READ
1MB
SEQ
READ
4KB
SEQ
WRITE
64KB
SEQ
WRITE
128KB
SEQ
WRITE
256KB
SEQ
WRITE
512KB
SEQ
WRITE
P411, 512MB, fw: 2.50, RAID 6 (256KB), D2700, 14-146GB SFF 6Gb SAS 15K, DL380-G6
P800, 512MB, fw: 7.00, RAID 6 (256KB), MSA70, 14-146GB SFF 3Gb SAS 15K, DL380-G5
ProLiant DL380 G6 servers with the Smart Array P411 and ProLiant DL380 G5 servers with the Smart
Array P800 were used for test results in Figures 3, 4 and 5. All other hardware and software
specifications are reported in Table 2.
Table 2. Hardware and firmware specifications for test results in Figures 3, 4, and 5
Test specifications
P411
P800
Firmware
v2.50
v7.00
Memory
512 MB
512 MB
Driver
HP Smart Array Controller Driver
HpCISSs2 (6.18.0.32 b14) RAID 0,5,6
(256KB)
HP Smart Array Controller Driver
HpCISSs2 (6.14.0.32 b25), RAID
0,5,6 (256 KB)
Internal disk
drives
1 logical drive
1 logical drive
External storage
D2700
MSA60
Drives
14-15K SFF SAS DP Seagate 146GB
6Gb (HPDB- firmware version)
14-15K LFF SAS DP Seagate 146GB
3Gb (HPDC- firmware version)
13
Data availability
Smart Array controllers support online array expansion, logical drive extension, stripe size migration,
and RAID migration. These technologies protect data and allow network administrators to modify the
array without interrupting user access. Smart Array controllers can monitor I/O activity, track key
parameters, predict potential problems, take corrective actions, provide automatic recovery, and
deliver full fault management to protect against downtime.
RAID support
In 1989, HP (then Compaq) was the first company to introduce RAID subsystems in the network server
marketplace. Today, RAID is an industry-standard technology and most online network data storage is
protected with some level of RAID.
Smart Array controllers support RAID levels 0, 1, 1+0, 5, 6, 50, and 60. Today, only Smart Array
controllers offer RAID 6, which provides higher fault tolerance than RAID 5 with lower implementation
costs than RAID 1+0 and greater usable capacity per unit density than RAID 1.
Stripe size migration
Using the ACU, administrators can change the stripe size of a logical drive without any downtime.
Stripe size migration can improve logical drive in response to changes in the read/write environment.
Each RAID level has a default value designed to provide good performance across many types of
applications. Table 3 shows the default values and the range of stripe sizes available for each RAID
level. Table 3 provides a generic overview of stripe sizes for the listed RAID levels and is generally
applicable to the present generation of Smart Array controllers. These values can change for certain
controllers and storage device implementations. Users should consult configuration documentation for
the specific controller and storage device in question.
Table 3. Stripe sizes available per RAID level
Fault-tolerance level
Default in KB
Available stripe sizes (KB)
RAID 0
128
8, 16, 32, 64, 128, 256
RAID 1 or 1+0
128
8, 16, 32, 64, 128, 256
RAID 5
64
8, 16, 32, 64, 128, 256
RAID 6 *
16
8, 16, 32, 64, 128, 256
RAID 50
128
8, 16, 32, 64, 128, 256, 512
RAID 60 *
16
8, 16, 32, 64, 128, 256
* Requires a valid SAAP license key
Certain applications, especially those performing mostly one type of transaction, write transactions for
example, may require adjusting stripe size to achieve optimal performance. Table 4 lists
recommended stripe sizes for general types of server applications. Administrators can start with these
general recommendations and then fine tune to determine the best overall performance for a
particular application. The ACU allows administrators to make these changes online without
disruption or data loss.
14
Table 4. Recommended stripe sizes
Type of server application
Suggested stripe size change
Mixed read/write
Accept default value
Mainly sequential read (such as audio/video
applications)
Larger stripe sizes
Mainly write (such as image manipulation
applications)
Smaller stripes for RAID 5, RAID 6
Larger stripes for RAID 0, RAID 1, RAID 1+0
NOTE:
Different controllers may have different stripe sizes. Users should consult the controller
user guide for more details.
RAID migration
Using the ACU, administrators can also change the RAID level of the logical drive without downtime.
Administrators can perform RAID migration to increase raw data storage capacity, improve
performance by increasing the number of spindles in a logical drive, or change fault tolerance (RAID)
configurations. Table 5 summarizes RAID levels and the amount of space required for each type of
fault tolerance.
Table 5. Summary of RAID methods
RAID 0
RAID 1
RAID 1+0
RAID 5
RAID 6
(striping)
(mirroring)
(striping and
mirroring)
Usable drive space*
100%
50%
50%
67% to 93%
50% to 96%
Usable drives
N
N/2
N/2
N-1
N-2
Minimum number of
drives
1
2
4
3
4
Tolerant of single drive
failure?
No
Yes
Yes
Yes
Yes
Tolerant of multiple
simultaneous drive
failures?
No
No
Yes, if failed
drives not
mirrored to
each other
No
Yes
Read performance
High
High
High
High
High
Write performance
High
Medium
Medium
Low
Low
Relative cost
Low
High
High
Medium
Medium
* The values for usable drive space are calculated assuming a maximum of 14 physical drives of the same
capacity (or a maximum of 56 for RAID 6) with no online spares. HP recommends not exceeding these
maximum figures (excluding any allowable online spares) when configuring a drive array due to the increased
likelihood of logical drive failure with more disk drives.
15
Drive roaming
HP Smart Array controllers support drive roaming, which allows administrators to move disk drives
and arrays while maintaining data availability. Drive roaming allows administrators to move one or
more disk drives that are members of a configured logical drive to a different bay position as long as
the new bay position is accessible by the same controller. In addition, it allows administrators to move
a complete array from one controller to another, even if controllers are in different servers.
Drive roaming is supported as an offline feature. There is no current support for “ejecting” an array
while the server is online and then importing it into a new physical location, whether it is attached to
the same array, or migrated to a separate online array.
Mirror splitting and recombining
Mirror splitting is a task that splits an array with one or more RAID 1 or RAID 1+0 logical drives into
two identical new arrays with RAID 0 logical drives. This is useful for administrators who want to
replicate a configuration or need to build a backup before performing a risky operation. Using the
ACU, administrators can also recombine a split mirrored array. Beginning with the present generation
of SAS-based Smart Array controller controllers, the Smart Array Advanced Pack (SAAP) is required to
use the mirror splitting and recombining functions. For support information regarding mirror splitting
on specific controllers, see the controller QuickSpecs. More information about mirror splitting can be
found in the technology brief “RAID 1(+0): breaking mirrors and rebuilding drives” at
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c00378986/c00378986.pdf
NOTE:
An array cannot be split if it contains logical drives in RAID 0, RAID 5, or RAID 6
configurations. An array can be split or re-mirrored only when the server is offline and
operating in the standard configuration mode of the ACU. When a split mirrored
array is recombined, all data on the second array is destroyed. For more information,
refer to the product documentation.
Online drive flash
Serial-based Smart Array controllers support online drive flashing, which saves time when updating
firmware. Instead of taking the hard disk drive (HDD) offline before loading a new firmware image,
administrators can download an updated HDD firmware image to the supported Smart Array
controller and update all of the HDDs the next time the server is rebooted. This greatly reduces the
time involved in updating disk drive firmware.
Recovery ROM
HP Smart Array controllers store a redundant copy of the controller firmware image to protect against
data corruption. If the active firmware image becomes corrupt, Smart Array controllers use the
redundant firmware image and continue operating. The recovery read-only memory (ROM) provides
protection against power outages during firmware flashing.
Pre-Failure Warranty using S.M.A.R.T technology
HP (then Compaq) pioneered failure prediction technology for disk disk drives by developing
monitoring tests run by Smart Array controllers. Called monitoring and performance (M&P), or drive
parameter tracking, these tests externally monitor disk drive attributes such as seek times, spin-up
times, and media defects to detect changes that could indicate potential failure.
16
HP worked with the disk drive industry to help develop a diagnostic and failure prediction capability
known as Self-Monitoring Analysis and Reporting Technology (S.M.A.R.T.). Over the years, as
S.M.A.R.T. matured, HP used both M&P and S.M.A.R.T. to support disk drive failure prediction
technology for Pre-Failure Warranty disk drive replacement.
S.M.A.R.T. has matured to the point that HP relies exclusively on this technology for disk drive failure
prediction to support Pre-Failure Warranty. Since 1997, all HP SCSI, SAS, and SATA server-class disk
drives have incorporated S.M.A.R.T. technology. S.M.A.R.T. disk drives have the capacity to inform
the host when a disk drive is experiencing abnormal operation likely to lead to drive failure.
S.M.A.R.T. improves failure prediction technology by placing monitoring capabilities within the disk
disk drive. These monitoring routines are more accurate than the original M&P tests because they are
designed for a specific drive type and have direct access to internal performance, calibration, and
error measurements. S.M.A.R.T. uses internal performance indicators and real-time monitoring and
analysis to improve data protection and fault prediction capability beyond that of the original M&P
tests. In addition, HP Smart Array controllers proactively scan the disk drive media during idle time
and repair, or report, any media defects detected.
S.M.A.R.T. can often predict a problem before failure occurs. HP Smart Array controllers recognize
S.M.A.R.T. error codes and notify HP Systems Insight Manager 5 (SIM) whenever a potential problem
arises. HP SIM, in turn, immediately notifies administrators of drive failures. The drive parameter
tracking feature allows Smart Array controllers to warn of potential drive problems before they occur.
HP drives that fail to meet the expected criteria may be eligible for replacement under the
HP Pre-Failure Warranty.
Automatic data recovery with rapid rebuild technology
When a disk drive in an array is replaced, Smart Array controllers use the fault-tolerance information
on the remaining drives in the array to reconstruct the missing data and write it to the replacement
drive. This process is called automatic data recovery, or rebuild. If fault tolerance is compromised, the
data cannot be reconstructed and is likely to be permanently lost.
The latest generation of HP Smart Array controllers includes rapid rebuild technology for accelerating
the rebuild process. Rapid rebuild technology comprises several features that result in substantially
quicker rebuild time. Faster rebuild time reduces the risk of logical drive failure by restoring logical
drives to full fault tolerance before a subsequent drive failure can occur.
Generally, a rebuild operation requires approximately 15 to 30 seconds per gigabyte for RAID 5 or
RAID 6. However, actual rebuild time depends on several factors, including the amount of I/O activity
occurring during the rebuild operation, the number of disk drives in the logical drive, the rebuild
priority setting, and the disk drive performance. The ACU allows administrators to view the rebuild
progress and set the priority for the rebuild operation.
Online spare
Smart Array controllers enable administrators to designate an unlimited number of drives as online
spares to arrays containing one or more fault-tolerant logical drives. The same spare drive can be
assigned to multiple arrays as a global spare. Smart Array configuration utilities ensure that SAS disk
drives can only be assigned as spares for SAS arrays (and likewise, SATA disk drives for SATA
arrays). During system operation, these spare drives remain up and running but not active; that is, no
I/O operations are performed to them during normal array operation. Spare drives are held in
reserve in case one of the active drives in the array fails, and then an online spare drive is selected as
the replacement disk drive.
5
For a more detailed discussion of HP Systems Insight Manager, refer to the “HP Systems Insight Manager” section.
17
If an active drive fails during system operation, the controller automatically begins rebuilding each
fault-tolerant logical drive onto the online spare; no administrator action is required. Once the rebuild
operation is complete, the system is fully fault-tolerant once again. The failed drive can be replaced at
a convenient time. Once an administrator installs a replacement drive, the controller will restore data
automatically from the failed drive to the new drive. At that point, the original online global spare, as
shown in Figure 6, will return to standby mode.
Figure 6. RAID 5 with an online spare drive
Smart Array Controller
User data
Parity data
Online spare
Dynamic sector repair
Under normal operating conditions and over time, disk drive media can develop defects caused by
variances in the drive mechanisms. To protect data from media defects, HP built a dynamic sector
repair feature into Smart Array controllers.
During inactive periods, Smart Array controllers configured with a fault-tolerant logical drive perform
a background surface analysis, continually scanning all drives for media defects. During busy
periods, Smart Array controllers can also detect media defects when accessing a bad sector. If a
Smart Array controller finds a recoverable media defect, the controller automatically remaps the bad
sector to a reserve area on the disk drive. If the controller finds an unrecoverable media defect and a
fault-tolerant logical drive is configured, the controller automatically regenerates the data and writes it
to the remapped reserved area on the disk drive.
ECC protection
HP Smart Array cache modules use ECC technology to protect cache data. DRAM is accessed by
transferring read-or-write data 32 or 64 bits at time, depending on the type of cache module. The
ECC scheme generates 8 bits of check data for every 32 or 64 bits of regular data. This check data
can be used not only to detect data errors, but also to correct them. Data errors could originate inside
the DRAM chip or across the memory bus.
Battery-backed write cache
HP Smart Array controllers ensure that data is protected in a location isolated from server failures
before acknowledging that the data transfer has completed. The write-back cache allows the
controller to acknowledge transfer completion before the data is physically stored in the disk drive. To
improve disk write performance, data is temporarily stored in battery-backed write cache (BBWC),
which uses DRAM and is substantially quicker when compared to the disk drive.
Battery power is required for RAID controllers to perform operations such as write-back cache, array
expansion, logical drive extension, stripe size migration, and RAID migration.
18
Recovering data from battery-backed cache
If an unexpected server shutdown occurs while data is held in BBWC, Smart Array controllers
automatically signal the memory chips to enter a self-refresh state and the controller initiates battery
power, or system auxiliary power if present. An amber LED, available either on the cache module or
battery pack, begins flashing to indicate that data is trapped in the cache. Smart Array controllers
automatically write this data to disk when power is restored to the system. If power is not restored
within the specified backup duration, the batteries may become drained of power. If that happens,
posted-write data in the cache will be lost. Once system power is restored, the batteries will
automatically recharge if needed. Battery recharge takes between 30 minutes and 2 hours,
depending on the remaining capacity level.
In the event of a server failure, the Smart Array controller and all of the drives can be moved to
another server to allow writing the data in the write cache to the drives.
In the event of a controller failure, the cache module containing posted-write data can be moved to a
new Smart Array controller. However, to preserve the cached data, the new Smart Array controller
must be attached to the original drives for which the posted-write data is intended.
Administrators should be aware of a special concern when using an embedded RAID controller with
battery-backed cache. If the server board fails, the replacement board must be the same model server
board so that the controller type and drive bays are the same. The cache module, battery pack, and
drives must be moved to the replacement system to extract the data from the battery-backed cache.
Selection criteria for battery-backed cache
HP Smart Array battery cells, battery enclosures, and contacts are custom designed to preserve the
integrity of business-critical information beyond the minimum specified backup duration. HP Smart
Array battery cells were selected to achieve the specified three-year backup life in typical server
environments.
A dedicated battery microcontroller continuously monitors the HP Smart Array battery pack for signs
of damage, including an open battery terminal, partial battery short, charge timeouts, and over
discharge conditions. Battery status information is indicated with an LED, power-on self-test (POST)
messages, event messages to the host, ACU information pages, ADU, and within HP SIM.
The battery microcontroller automatically disables the battery-backed cache features any time it
detects battery damage or the charge level falls below the required limits to achieve the specified
backup duration. The battery microcontroller automatically restores battery-backed cache features
when the microcontroller detects a replacement battery or when battery recharging is complete. Highend HP RAID controller designs contain two batteries to protect against a single battery cell failure
while data is held in cache.
For detailed technical information on all HP cache options and controller compatibility, go to
www.hp.com/products/smartarray.
Types of batteries
HP Smart Array controllers use rechargeable Nickel Metal Hydride (NiMH) button cell batteries
specifically designed for longer life at the temperatures found inside rack-mounted servers. Typical
capacity reduction for the HP Smart Array battery pack is 5 to 10 percent over a 3-year period,
depending on server temperature and number of discharge cycles.
NiMH cells are also environmentally friendly, since they do not contain harmful lead, mercury, or
cadmium material. Additionally, NiMH chemistry does not suffer capacity memory effects that can
lower battery capacity. For example, memory capacity of Nickel Cadmium (NiCD) batteries is
reduced when the batteries are exposed to short discharge cycles. Lithium Ion (Li-Ion) batteries are
typically smaller than NiMH batteries, but their capacity is permanently reduced at high temperatures
and they are usually limited to 100 full discharge cycles.
19
The HP 650 mAh P-Series battery has the same form factor as previous versions and extends battery
life up to 48 hours before recharging is necessary.
Battery replacement
HP Smart Array controllers include serviceable battery packs that allow tool-free battery pack
replacement with no need to replace either the Smart Array controller or the detachable cache
module.
Alternatives to battery replacement
If it is not possible or desirable to replace the batteries, administrators have three options to disable
write-back cache and avoid losing critical data:
1. The ACU provides a method for adjusting the read-and-write cache ratio to 100 percent read
cache.
2. The ACU provides a method to disable the array accelerator for each logical drive, which disables
both read-ahead and write-back cache.
3. Administrators can replace an existing RAID controller with a newer Smart Array controller model.
NOTE:
If the write cache is turned off, some write performance degradation may occur.
For more information about Smart Array controller battery configuration and specifications download
the “HP Smart Array Controllers for HP ProLiant Servers User Guide”:
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01608507/c01608507.pdf?ju
mpid=reg_R1002_USEN
Flash-backed write cache
HP introduced the flash-backed write-cache (FBWC) system in the fourth quarter of 2009 (Figure 7).
20
Figure 7. HP flash-backed write-cache and Super-cap power supply
The FBWC uses NAND 6 flash devices to retain cache data and super-capacitors (Super-caps) instead
of batteries to provide power during a power loss. The FBWC offers significant advantages over the
HP Battery-backed write-cache (BBWC) system. Since the FBWC writes the contents of memory to
flash devices, there is no longer a 48 hour battery life limitation and the data will be posted to the
disk drive on the next power up.
FBWC architecture
The FBWC DDR2 mini-DIMM cache module is specifically designed for the present generation of
PCIe2.0, SAS-based Smart Array controllers based on the PMC PM8011 max SAS SRC 8x6G RAID
on a chip (RoC). The primary FBWC components consist of the cache module, Super-caps with
integrated charger, and a RoC located on the system board, shown in Figure 8.
Non-volatile semiconductor memory that can be electronically erased and reprogrammed. No power is needed to maintain data stored in the
chip
6
21
Figure 8. FBWC block diagram
Side band
control
signals
NAND Flash
4b 33MHz
4b 33MHz
Data
FPGA
Command
& address
Super-cap
NAND Flash
In off-module pack connecting to
cache module
PROM
DRAM
8X
DRAM
8X
DRAM
8X
133 MHZ DDR IF
Register
Cache module
System board
400 MHZ DDR IF
Cache dirty N*
Reset N
Reg reset N
RoC
TWI**
* Cache tracks that have been written over are designated as "dirty"
** Two wire interface (TWI)
FBWC cache
The FBWC cache module with a field programmable gate array (FPGA), DDR2 DRAMs, and NAND
flash devices can support up to 1 GB of DDR2 memory and up to 72 data bits (64 data bits plus 8
ECC bits). The FBWC can support up to 800 Mbps data rate when the Smart Array controller is
driving the DDR2 bus. When the FPGA is driving the bus in a data recovery situation, the data rate is
266 Mbps. The FBWC module connects to the Smart Array controller through a 244-pin mini-DIMM
connector.
At the time of publication, the FBWC cache is supported on the Smart Array 410i with support for
other present generation Smart Array controllers to follow in the first quarter of 2010.
Super-capacitor
The Super-cap module sub-assembly consists of two 35 Farad 2.7V capacitors, configured in series,
providing 17 Farads at up to 5.4V. The charger maintains the Super-cap at 4.8V, providing the
required amount of power to complete backup operations while extending the life of the Super-cap.
The charger also monitors Super-cap health and activates LED indicators on the FBWC module to
warn of impending failure. The Super-Cap module is contained within the same form factor and
housing as the HP 650 mAh P-Series battery used in the HP BBWC.
Capturing data during power loss
Loss of power in a server using the FBWC prompts the FPGA to copy data contained in the DRAM to
the NAND flash devices residing on the cache module. The Super-caps supply the energy needed to
power the FBWC system while performing the data backup operation.
22
Recovering data from the flash-backed cache
When system power is present, the FPGA on FBWC is in its idle state. In the idle state, the FPGA
simply monitors the voltage statuses, the resets, and the control signals managed by the Smart Array
controller. The FPGA’s DDR2 I/O pins are held in “tri-state,” equivalent to a disabled mode, to avoid
bus contention. When system power is lost, the FPGA waits for the Smart Array controller’s clock
enable signal to transition to low, signaling that the controller has stopped driving the DDR2 bus. At
this time, the FPGA assumes control of the bus and begins moving data from the DRAMs to the nonvolatile flash memory. Upon the next power up, the FPGA then restores the cache by moving data
from the flash memory to the DRAMs.
Once cache has been restored to the DRAMs, the Smart Array controller verifies that the data has
been retained correctly. If so, the data is transferred to the disk disk drives.
Entry level RAID solutions
Software RAID
HP has developed a software RAID solution based on the Smart Array firmware. The B110i SATA
Software RAID supports the ACU, ACU-CLI (command line tool), SNMP agents, and Web-Based
Enterprise Management (WBEM) providers.
The B110i features an OS-specific driver from HP that uses the embedded ICH10R controller. The
B110i supports RAID 0, 1, and 1+0 and a maximum of two logical drives. The B110i supports up to
four 1.5G or 3G SATA drives. Administrators can move drives to a Smart Array controller in a
seamless procedure that maintains the user data and RAID configuration.
The ProLiant DL320 G6 server supports the B110i. For a listing of the complete feature set and
support information for the B110i SATA Software RAID, download the B110i user guide at
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01706551/c01706551.pdf
Zero Memory RAID
Using Zero Memory RAID (ZMR), administrators can create a RAID 0-1 configuration without
additional memory. ZMR uses memory embedded in the controller, approximately 1K in size, and
supports limited configurations. ZMR supports up to eight drives in Zero Memory Mode, or seven
drives and one tape drive. ZMR does not include caching; however, all systems can be upgraded to a
BBWC memory module that can significantly increase performance.
ZMR is supported on present generation Smart Array controllers for internal, direct connections only.
This includes the Smart Array P410, P410i, P212, and P712 controllers. Consequently, the P212
controller does not include ZMR on the external connector. Modular Smart Array (MSA) products are
not supported in ZMR mode.
NOTE:
The P212 controller can only be upgraded to 256 MB BBWC, so it does not support
512 MB BBWC.
NOTE:
Smart Array Advanced Pack is not available on Zero Memory configurations.
23
Storage support and pathway redundancy
HP Smart Array controllers support solid state drives, Native Command Queuing and Dual Domain
providing increased performance and redundancy on the storage network. Smart Array controllers
continue to support tape back up devices.
Solid state drives
HP has introduced the second generation of solid state drives (SSD) for ProLiant servers. These solid
state drives are 3Gb/second SATA interface in both 60GB and 120GB capacities. The product,
based on NAND Single Level Cell flash technology, are implemented as SFF and LFF hot plug devices
on the HP universal drive carrier for general use across the ProLiant portfolio. These drives deliver
higher performance, lower latency, and low power solutions when compared with traditional rotating
media.
These HP second generation SSDs are supported for use with the present generation Smart Array
controllers based on the PM8011 SRC MIPS processor on select ProLiant G6 servers.
Native Command Queuing
Native Command Queuing (NCQ) increases SATA HDD performance by internally prioritizing
read/write command execution. This reduces unnecessary drive head movement which results in
increased performance, especially in server or storage-type applications where multiple simultaneous
read/write requests are common. Without NCQ, the drive can process and complete only one
command at a time. NCQ must be supported in both the controller and the drive.
NOTE:
For more information about NCQ support on Smart Array controllers, see Support
Communication – Customer Notice, Document ID: c01800086, Version: 1 athttp://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=c01
800086&lang=en&cc=us&taskId=101&prodSeriesId=428936
Dual Domain support
Dual Domain SAS creates redundant pathways from servers to storage devices. These redundant
paths reduce or eliminate single points of failure within the storage network and increase data
availability. Dual domain SAS implementations can tolerate host bus adapter (HBA) failure, external
cable failure, expander failure, or failure in a spanned disk (JBOD).
Dual Domain support is available for the HP Smart Array P800 attached to an MSA60/70 with the
HP StorageWorks Dual Domain I/O Module Option (AG779A). Dual Domain support requires HP
Smart Array firmware v5.10 or higher and dual-port SAS drives.
Dual Domain supports Multi-Initiator JBOD Clusters. I/O Module Option with the HP SC44Ge Host
Bus Adapter supports HP-UX and OpenVMS on select HP Integrity servers only.
Dual Domain is also supported by HP Serviceguard software on HP-UX with HP-UX 11iv2 and 11iv3
only.
24
Tape device support
Smart Array controllers support tape back-up devices. The One Button Disaster Recovery (OBDR)
ProLiant server/controller compatibility matrix for currently shipping HP products is available at
www.hp.com/go/obdr.
Smart Array Advanced Pack
HP Smart Array Advanced Pack (SAAP) firmware provides advanced functionality within Smart Array
controllers. This firmware further enhances performance, reliability, and availability of data. SAAP is
hosted on the Smart Array controller hardware firmware stack. It can be enabled beginning with the
present generation of Smart Array controllers.
SAAP requires a license key for activation. After activation, administrators can use several standard
capabilities:
• RAID 6 protects against failure of any two drives. It requires a minimum of four drives, but only two
will be available for data. RAID 6 can tolerate multiple simultaneous drive failures without downtime
or data loss and is ideal for applications requiring large logical volumes, because it can safely
protect a single volume of up to 56 disk drives. RAID 6 also offers lower implementation costs and
greater usable capacity per U than RAID 1.
• RAID 60 (RAID 6+0) is a nested RAID method that uses RAID 0 block-level striping across multiple
RAID 6 arrays with dual distributed parity. RAID 60 allows administrators to split the RAID storage
across multiple external boxes. It requires a minimum of eight drives, but only four will be available
for data.
NOTE:
HP BladeSystem servers do not support RAID 6 and 60
• Advanced Capacity Expansion (ACE) automates higher capacity migration using capacity
transformation to remove logical drives by shrinking and then expanding them online. Standard
drive migration and expansion remain unchanged.
• Mirror Splitting and Recombining in Offline Mode breaks a RAID 1 configuration into two RAID 0
configurations. This is similar to a scaled down rollback functionality that requires two disk drives.
• Drive Erase completely erases physical disks or logical volumes. This capability is useful when
decommissioning, redeploying, or returning disk drives.
• Video On Demand Performance Optimization optimizes performance of video on demand and
improves latency during video streaming.
More information about SAAP is available at www.hp.com/go/SAAP.
NOTE:
At a minimum, a 256 MB cache and battery kit is required to enable the
SAAP license key. SAAP is not available on Zero Memory Configurations.
25
Storage management
To improve data storage management, HP Smart Array controllers include built-in intelligence that
makes it easier for administrators to configure, modify, expand, manage, and monitor storage.
HP provides five utilities for managing an array on a Smart Array controller:
• ACU (Array Configuration Utility)—ACU is the main tool for configuring arrays on HP Smart Array
controllers. It exists in three interface formats; the ACU GUI, the ACU CLI, and ACU Scripting. All
three formats have separate executables:
– Starting with version 8.28.13.0, ACU Scripting is now a standalone application that is
distributed with the ACU CLI application. In ACU versions prior to 8.28.13.0, the scripting
executable was provided with the ACU GUI component.
– Users familiar with the previous versions of ACU Scripting must now install the ACU CLI
application to obtain the scripting executable. The new ACU scripting executable (hpacuscripting)
replaces the former executable (cpqacuxe) in all scripts.
• ORCA (Option ROM Configuration for Arrays)—A simple ROM-based configuration utility
• CPQONLIN—A menu-based configuration utility specifically for servers using Novell NetWare
• HP System Insight Manager—Extensible, secure management suite for server and storage
environments
• ADU (Array Diagnostic Utility)—A diagnostic and reporting utility for Smart Array controllers. With
the introduction of version 8.28.13.0 or later of ACU or HPACUCLI, ADU is no longer provided as
a separate utility but now has its functionality integrated with ACU and HPACUCLI.
For more information about storage management and using these utilities, see the “Configuring
Arrays on HP Smart Array Controllers Reference Guide,” available at
http://h20000.www2.hp.com/bc/docs/support/SupportManual/c00729544/c00729544.pdf
Table 6 summarizes the capabilities of ACU, ORCA, and CPQONLIN utilities.
Table 6. Summary of HP configuration utility features
Utility features
ACU
ORCA
CPQONLIN
Uses a graphical interface
Yes
No
No
Available in languages other than English
Yes
No
No
Executable at any time
Yes
No
Yes
Available on disc
Yes
No
No
Uses a wizard to suggest the optimum
configuration for an unconfigured controller
Yes
No
Yes
Describes configuration errors
Yes
No
No
Creating and deleting arrays and logical drives
Yes
Yes
Yes
Assigning RAID level
Yes
Yes
Yes
Sharing spare drives among several arrays
Yes
No
Yes
Assigning multiple spare drives per array
Yes
No
Yes
26
Utility features
ACU
ORCA
CPQONLIN
Setting stripe size
Yes
No
Yes
Migrating RAID level or stripe size
Yes
No
Yes
Configuring controller settings
Yes
No
Yes
Expanding an array
Yes
No
Yes
Creating multiple logical drives per array
Yes
No
Yes
Setting of boot controller
No
Yes
No
Array Configuration Utility
The ACU 7 is a browser-based graphical application that helps configure Smart Array controllers. The
ACU is also supported on the HP MSA family of entry-level SAN products. This provides
administrators with a seamless set of tools to use with both HP Smart Array direct-attached storage
(DAS) and HP SAN-attached storage.
The ACU runs online on Microsoft® Windows® Server and Linux® operating systems. Because the
ACU is available on a bootable disc, an administrator using other operating systems 8 can run the
utility offline by booting the system from the SmartStart or Server Storage Support Software disc.
The ACU also contains a command line interface (ACU-CLI), offering a quicker way to deploy multiple
servers by automating creation of arrays and logical drives. Configuration information for the ACU is
available at
http://bizsupport1.austin.hp.com/bc/docs/support/SupportManual/c00729544/c00729544.pdf
Option ROM Configuration for Arrays
Option ROM Configuration for Arrays (ORCA) is an alternative method of viewing, creating, and
deleting multiple arrays and logical drives during system power up. ORCA is designed for
administrators who have minimal configuration requirements. The ACU is recommended for more
advanced array configurations.
CPQONLIN
CPQONLIN is a configuration utility that runs online on Novell NetWare. It functions like the ACU in
this environment. Refer to the HP Smart Array controller documentation for more information about
using CPQONLIN.
HP Systems Insight Manager
HP Systems Insight Manager is a client/server tool for integrated server environment management.
Based on SNMP, it is capable of monitoring more than 1,200 system-wide parameters for
performance and other operational characteristics of Smart Array controller storage. The program
displays configuration information, operating system device driver version numbers, controller
firmware version numbers, Pre-Failure Warranty information, and operating statistics.
For more information about the ACU, refer to
http://h18004.www1.hp.com/products/servers/proliantstorage/software-management/acumatrix/index.html.
8
Refer to product documentation to verify operating system support.
7
27
Performance monitoring
HP SIM gives administrators a window to look at low-level performance characteristics of Smart Array
controllers in the environment. It monitors three basic Smart Array controller performance parameters
for proactive storage subsystem management:
• I/O commands per second
• Average command latency
• Local processor utilization
Analyzing these key parameters can assist administrators in fine tuning configurations for
performance. HP SIM can graphically chart performance over time for each of these parameters.
Fault prediction
A background task monitors several key drive parameters and notifies HP SIM if a drive fails to meet
certain factory-preset criteria. HP SIM alerts the administrator to the potential problem.
Array Diagnostics Utility
The HP Array Diagnostics Utility (ADU) is an in-depth diagnostic and reporting utility for all Smart
Array controllers. The ADU quickly identifies problems such as incorrect versions of firmware,
improperly installed drives, inappropriate error rates, and failed batteries on the array accelerator
board.
The ADU displays a detailed analysis of the system configuration. If the cause is not apparent, the
ADU can generate a full report for administrators to fax or e-mail to HP customer service for phone
support.
Summary
HP Smart Array controllers are powerful I/O solutions for today’s most demanding network data
storage requirements. Smart Array controller technology combines an advanced feature set and fullspectrum, hardware-based fault management with exceptional performance characteristics.
Smart Array controllers provide solutions for all four primary data storage requirements: capacity
growth, high performance, data availability, and manageability. Smart Array controllers represent
patented HP technology and continue the HP commitment to providing the most complete and
advanced storage solutions for server customers.
HP is the only server provider with a seamless storage solution set that spans the range from
embedded Smart Array controllers in servers to plug-in PCI Smart Array controllers to SAN-attached
MSA storage. The tools used for managing and configuring storage are the same in all of those
environments. Data sets are compatible across all of those environments. The HP universal drive
strategy allows customers to easily migrate data from DAS to SAN.
28
Appendix A: Capacity growth technologies
The Smart Array controller family includes a standard toolset that administrators can use to configure
array controllers, expand an existing array configuration by adding disk drives, or reconfigure an
array by extending logical drive sizes. Before this innovation, expanding the storage capacity
attached to an array controller required a time-consuming backup-reconfigure-restore cycle.
Array expansion
Array expansion is the process of adding physical drives to an array that has already been
configured. The logical drives that exist in the array before the expansion takes place are unchanged;
only the amount of free space in the array changes.
For example, suppose an existing array consists of four physical drives and the administrator wants to
expand the array to six physical drives. This is like having four glasses full of water and dividing that
water from the original four glasses among six glasses (Figure A1). The amount of water (the size of
the logical drive) has not changed—it has merely been redistributed, or expanded, into a larger
number of containers (drives).
Figure A1. Array expansion redistributes an array into a larger number of physical drives. The size of the logical
drive does not change.
Thus, if an existing array is nearly filled with logical drives, an administrator can add new physical
drives and initiate an array expansion using the ACU. The ACU automatically checks the drive
hardware configuration to ensure the array expansion. Then, the existing logical drive is distributed
across all physical drives in the expanded array without affecting any existing data. If the array being
expanded contains more than one logical drive, data is redistributed one logical drive at a time.
The expansion process is entirely independent of the operating system. For example, if a
10-gigabyte (GB) logical drive is expanded from four drives to six, the operating system is unaware
of the change.
The amount of time it takes to complete an online array expansion depends on subsystem I/O activity,
the size of the logical drives on the array, fault-tolerance level, the number of disk drives in the new
array, disk drive performance, and the priority level of the array expansion. Administrators can
monitor array expansion completion progress within the ACU.
29
The ACU allows administrators to set the priority of array expansion operations. A high-priority setting
prioritizes the array expansion operation over I/O requests. A low-priority setting is the ACU default
setting and gives I/O commands precedence over array expansion. During idle time, when no I/O
commands are active, the array expansion operation runs at full speed regardless of the priority
setting.
Logical drive creation
Once the array capacity is expanded, the added capacity can be used to create a new logical drive
(Figure A2) or to extend the size of an existing logical drive (described in the “Logical drive
extension” section).
Figure A2. After array expansion, the administrator can use the free space by creating a new logical drive.
For example, if every department in a company has its own logical drive and a new department is
created, the administrator might need to create an entirely new logical drive for that department.
However, many administrators wish to extend the size of their logical drive after array expansion.
Logical drive extension lets administrators increase logical drive size if the existing drive is running out
of storage space.
30
Logical drive extension
Logical drive extension increases the storage space of a logical drive (Figure A3). During this process,
an administrator adds new storage space to an existing logical drive on the same array. An
administrator may have gained this new storage space either by array expansion or by deleting
another logical drive on the same array.
Figure A3. Logical drive extension grows the size of a logical drive.
The operating system must be aware of changes to the logical drive size. Refer to operating system
documentation for additional information.
31
For more information
For additional information, refer to the resources listed below.
Resource description
Web address
Smart Array controllers
www.hp.com/products/smartarray
HP disk disk drive products
www.hp.com/products/harddiskdrives
HP Modular Smart Array external
storage systems
www.hp.com/go/msa
HP ProLiant servers
www.hp.com/go/proliant
HP SAS Technology
www.hp.com/go/serial
Configuring Arrays on HP Smart Array
Controllers Reference Guide
http://h20000.www2.hp.com/bc/docs/support/SupportManual/
c00729544/c00729544.pdf
Smart Array Advanced Pack
www.hp.com/go/SAAP
Serial Attached SCSI storage technology
brief
http://h20000.www2.hp.com/bc/docs/support/SupportManual/
c01613420/c01613420.pdf.
Redundancy in enterprise storage
networks using dual-domain SAS
configurations
http://h20000.www2.hp.com/bc/docs/support/SupportManual/
c01451157/c01451157.pdf
HP Smart Array controllers for HP
ProLiant Servers User Guide
http://h20000.www2.hp.com/bc/docs/support/SupportManual/
c01608507/c01608507.pdf?jumpid=reg_R1002_USEN
HP One-Button Disaster Recovery
(OBDR) Solution for ProLiant
http://h18006.www1.hp.com/products/storageworks/drs/index.h
tml
Call to action
Send comments about this paper to TechCom@HP.com.
© 2004, 2006, 2007, 2008, 2009 Hewlett-Packard Development Company,
L.P. The information contained herein is subject to change without notice. The
only warranties for HP products and services are set forth in the express
warranty statements accompanying such products and services. Nothing herein
should be construed as constituting an additional warranty. HP shall not be
liable for technical or editorial errors or omissions contained herein.
Microsoft and Windows are U.S. registered trademarks of Microsoft
Corporation.
Linux is a U.S. registered trademark of Linus Torvalds.
TC091204TB, December 2009
32
Download PDF