2. A Blueprint for Lowering Capital
and Operational Costs for
Storage Infrastructure
Unprecedented data growth, new application demands, regulatory
requirements for data archival and greater business continuity
needs are straining IT budgets and exposing inefficiencies in
conventional storage architectures. One of the most pressing
issues facing organizations today is the need to reduce storage
costs. Simultaneously, single vendor management tools and
isolated pools of storage fragment the infrastructure, making it
difficult to share resources, improve operational efficiency and
migrate online data. By implementing a virtualized tiered storage
architecture, an IT organization can dramatically improve storage
capacities and lower capital and operational expenses. Virtualized
tiered storage better aligns data on storage systems by allowing
you to efficiently match storage attributes with the service level
needs of individual business applications.
legislation, are setting mandatory guidelines
The Cost and Complexity
for data protection and data retention, with
of Storage
financial and legal penalties for noncompli-
Storage environments large and small face ance. Like business resilience measures,
increasingly sophisticated demands from ensuring your organization is in compliance
business applications and requests to with government regulations implies
maintain data online for longer periods of time. increased capacity and complexity in the
As a result, total capacity being managed by storage environment. Meeting the need for
IT storage administrators has exploded and enhanced data protection, greater regulatory
the complexity of the storage infrastructure compliance and escalating demand for
has grown enormously. These factors are application storage capacity inevitably drives
raising the total cost of the storage ownership up the cost of storage.
and significantly affecting the IT budget.
Hardware Costs
Regulatory compliance efforts offer an
Fragmentation of the storage environment,
example. The Health Insurance Portability
caused by disconnected islands of storage
and Accountability Act (HIPAA), SEC rule
and interoperability problems, creates
17a-4, Sarbanes-Oxley and New Basel
stranded capacity and duplicates storage
Capital Accord (Basel II), along with other
networking equipment, contributing to
3. escalating hardware costs. Eliminating barriers
to sharing, storage recovery and improving
capacity utilization rates can deliver sustainable
long term savings by allowing future purchase
of storage assets to be deferred.
Software Costs
With heterogeneous configurations the norm
in enterprise data centers, managing storage
requires the use of many software tools from
different vendors. These tools frequently
do not communicate well with each other,
complicating the process of provisioning,
optimizing, moving and protecting data and
increasing training costs.
labor Costs
Labor expenses represent as much as
40 percent of the total cost of storage
ownership. The best meaningful way to NAS = network attached storage
lower storage labor costs is to reduce the CAS = content addressable storage
VTL = virtual tape library
operational management complexity of the
storage environment. Figure 1. The Hitachi Universal Storage Platform® V (or Hitachi Universal Storage
Platform VM) virtualizes a heterogeneous storage environment.
Maintenance Costs
Hardware and software maintenance costs
correlate directly with the amount of capacity
Mission Critical Data
deployed in the environment and the number ■■ Most valuable to an enterprise, high access
of tools in use by storage administrators. ■■ High performance, high availability, near zero downtime, highest cost
Improvements that simplify storage admin-
istration and reduce the number of tools Business Critical Data
required also benefit maintenance costs. ■■ Important to the enterprise, average cost
■■ Reasonable performance, good availability, less than eight hour recovery
Soft Costs
Soft costs include storage system perfor- Accessible online Data
■■ Cost sensitive, low access, large volumes
mance, scalability, downtime and regulatory
compliance. For example, if inadequate data
■■ Online performance, high availability, less than eight hour recovery
protection raises the risk of revenue and nearline Data
productivity losses from an outage, actual ■■ Cost sensitive, low access, large volumes
losses experienced during downtime are ■■ Less than one hour access time, automated retrieval
attributable to the storage infrastructure, even
if the failure occurred elsewhere. offline Data
■■ Archived data, backup or compliance related
Very cost sensitive, limited access, ~72-hour seek time
Solution: Virtualized
■■
Tiered Storage Table 1. IDC Tiered Storage Data Classifications
Today's enterprise data centers feature storage
systems and storage network infrastructures enable the separation of logical views from can strategically align storage infrastructure
from many vendors. In order to support this, physical assets, enabling a Services Oriented with business requirements.
the Hitachi virtualized tiered storage archi- Storage Solutions architecture that provides
E-mail offers an excellent example of how
tecture consolidates heterogeneous storage the right cost, performance, reliability and
tiered storage architectures lower costs.
solutions in a single managed pool (see Figure availability characteristics of storage, as
Not all e-mail data has the same value.
1). Within this pool different storage tiers are needed, matched with application require-
Typically, only relatively recent messages
established. Hitachi virtualization technologies ments (see Table 1). This means organizations
have the high end requirements associated
4. with top-of-the-line storage systems. As
Tier Name Mission Business Accessible Nearline Offline
messages age and are archived, access fre-
Critical Critical Online
quency declines and needs change. Storing
Availability 99.999% 99.999% 99.99% 99.90% Offline
all e-mail data on high-performance, highly
Performance 5 4 3 2 1
available storage is wasteful.
Maximum Backup Time (hours) 0.02 0.02 3 0.02 N/A
In a tiered storage environment, the IT ad- Maximum Local Recovery Time 1 4 6 4 1
ministrator can establish policies where only (hours)
new e-mail messages are stored on tier one, Local Recovery Point (hours) 24 24 24 24 N/A
high end, enterprise class storage. Older Remote Recovery Time (hours) 1 72 72 72 72
e-mail messages can be hosted on tier two Remote Recovery Point (days) 1 7 7 30 N/A
storage, offering performance and availability Offering Cost/GB (relative) 100% 80% 60% 40% 10%
slightly below tier one at a much lower cost. Point-in-time Snapshots Yes Yes No No No
And archival messages, the bulk of all e-mail
storage, can be stored on cost efficient Table 2. Metrics of a Tiered Storage Implementation
midrange systems using SATA disk, offering
online access but with lesser performance. ■■ Seamless, nondisruptive data migration Hitachi Basic operating System V
This tiered configuration frees large amounts between heterogeneous storage The Hitachi Basic Operating System V soft-
of high end storage capacity for use by resources without affecting access to ware package is an upgrade to the Hitachi
other applications and lowers the cost of data by business applications Basic Operating System that adds the
providing storage for the e-mail application solution’s unique external storage virtualiza-
(see Table 2). Key Components tion capabilities to the Universal Storage
Platform V and Universal Storage Platform
The Hitachi virtualized tiered storage archi- Hitachi virtualized tiered storage solutions
VM. It also expands the virtual storage parti-
tecture leverages superior Hitachi controller are designed to provide government,
tion support to 32. These capabilities enable
technology to provide thin provisioning, enterprise and midsize businesses with
organizations to take advantage of Hitachi
advanced replication, logical partitioning, scalable, cost effective storage solutions
storage management, replication and data
virtual connections, support for heteroge- with heterogeneous storage support, tier
mobility capabilities across an entire hetero-
neous storage and the sheer performance oriented provisioning and built in nondis-
geneous storage pool and add significant
and scalability needed to truly deliver stor- ruptive data mobility. Built around Hitachi
value to their Universal Storage Platform V
age services on demand — while at the Universal Storage Platform® V and Hitachi
or Universal Storage Platform VM.
same time eliminating vendor lock-in. Tiered Universal Storage Platform VM storage
storage can be scaled out using hardware systems, and complementary to the Hitachi Hitachi tiered Storage Manager
appropriate to the tier. This dramatically NAS Platform, powered by BlueArc®, as According to the Enterprise Strategy Group,
simplifies storage provisioning and manage- well as content archiving and data resilience data migration is a frequent task performed
ment and helps organizations achieve: products from Hitachi Data Systems, prod- by data center storage administrators.
ucts specific to virtualized tiered storage Whether managing the replacement, reloca-
■■ Management simplicity through the use solutions include: tion, consolidation or lease renewal of serv-
of common storage services throughout
ers and storage or balancing workloads for
an entire heterogeneous storage Hitachi Basic operating System
performance, data migration is a fact of life
infrastructure The Hitachi Basic Operating System
for the storage administrator. Data migration
■■ A radically more efficient IT environment software package is used to configure and
is also a fundamental component of a tiered
that goes beyond consolidation into manage Universal Storage Platform V and
storage strategy. As the cost, quality of
dynamically allocated tiered storage Universal Storage Platform VM systems.
service and functionality demands of appli-
It supports virtualizing front end controller
■■ Asset value protection by supporting the cation data change over time, data must be
ports and dynamic partitioning of cache and
virtualization of heterogeneous storage migrated between tiers. This is how a tiered
system resources through up to four virtual
systems from leading vendors storage architecture maintains an optimal
storage partitions. It also includes basic
■■ Storage reclamation and improved match of storage system characteristics
reporting and performance monitoring and
utilization of physical assets and data requirements.
management tools.
5. Hitachi Tiered Storage Manager data man-
Virtualized Tiered Storage Benefits
agement software enables IT administrators
to easily and interactively match application
quality of service requirements to hetero- Business Benefits
geneous storage assets. Tiered Storage ■■ Capital expense savings and lower operational costs
Manager enables easy storage tier creation
■■ Centralized management and better storage visibility
and the nondisruptive movement of data
volumes to match application driven price,
■■ Increased utilization
performance and availability characteristics. ■■ Higher availability
■■ Reduced application service interruptions
Hitachi Dynamic Provisioning
■■ Storage infrastructure aligned with business requirements
For companies faced with ongoing rapid
growth of their data storage requirements technical Benefits
and escalating storage and storage manage- ■■ Simplified provisioning and storage management
ment expenses, Hitachi Dynamic Provision- ■■ Precise storage service level management
ing software uses thin provisioning to simplify
■■ Leverage of lower cost storage resources
and add agility to the application storage
■■ Nondisruptive data migration
provisioning process and can save significant
money on storage purchases. It also can
■■ New functionality extended to older assets
reclaim existing previously allocated storage ■■ Enhanced performance and automated performance optimization
within the virtualized pool — including on ■■ Thin provisioning
externally attached heterogeneous systems. ■■ Improved security from logical partitioning, role-based access, audit logging
And to improve performance and simplify and other security features
performance management, Hitachi Dynamic
Provisioning software automates optimiza-
tion of spindle usage and provides dynamic
load balancing.
the different tools. The fragmented manage- The storage controller-based virtualization
ment tools subsequently resulted in siloed of the Universal Storage Platform allowed
University of Utah Sees management practices and in administra- UUHSC to implement a new tiered stor-
Success and a Positive tors themselves becoming a single point age architecture. All fabric elements were
Return on Investment of failure for the environment. kept the same, older storage systems were
The University of Utah Health Sciences virtualized behind the Universal Storage
In addition to storage management con-
Center (UUHSC) Information Technology Platform, SAN management was collapsed
cerns, backup windows started to intrude
Services (ITS) group provides technical into a single point of control and a multi-
on production work hours, and the amount
resources and support for a wide range of tiered storage strategy was implemented.
of storage capacity wasted by stranded
departmental, hospital, research group and
space on SANs was growing. economic impact of the Hitachi
clinic-based facilities in Salt Lake City, Utah.
universal Storage Platform
With demand for storage capacity growing Storage virtualization was identified as
The implementation of the Hitachi Universal
at around 200 percent per year, and more a technology that could address many
Storage Platform and a tiered storage
UUHSC departments signing up to share UUHSC growth and management issues.
infrastructure resulted in a dramatic range
resources managed by the ITS group, the However, after ITS evaluated several appli-
of financial savings for the UUHSC:
existing storage infrastructure was begin- ance and fabric-based storage virtualization
ning to show signs of stress. products, it determined that the risk of man- ■■ Reduced the number of people needed
aging another fabric element, the neces- to manage the storage infrastructure by
ITS supports several SANs from a num-
sary increase in head count and training, two (saving US$320K)
ber of different vendors. But, as deployed
and the single point of failure that would be ■■ Improved capacity use, deferring planned
capacity increased, storage administrators
introduced into the system did not warrant purchase of 18TB of new capacity for
noted several problems. Each vendor's SAN
further investigation. The ITS group then three years
requires its own set of tools and storage
decided to evaluate the Hitachi Universal
management procedures. This frustrated
Storage Platform.
attempts by administrators to cross train on