subscribe: Daily Newsletter

 

The importance of data centre transformation

0 comments

In today's connected global marketplace, data centres are more important than ever. Customers expect to purchase products and access services 24×7 – so any downtime, planned or unplanned, can disrupt critical business processes.


New regulations are requiring organisations to store larger quantities of data for longer periods of time. And environmental concerns mean data centres are under increasing pressure to reduce their carbon footprint and dispose of old equipment in a safe and eco-friendly manner.
To meet these evolving demands, data centres must achieve near-continuous availability, increase capacity and efficiency, and become greener.
But not all data centres are ready for these challenges.
Many large enterprises have multiple data centres that are 15 years old, running multiple servers and operating platforms.
Unused capacity, redundant functionality, inefficient or outdated designs, an ever-growing number of assets, and aging servers make these environments complex and expensive to manage and maintain as well as difficult to scale. In addition, power costs are rising steadily, leading to exorbitant electric bills.
Given all these factors, key questions for data centre managers today include:
* Are our data centres ready to enable the business growth?
* Can our data centres support the business globally, 24×7?
* Should we consolidate data centres, build new ones, or modernise existing sites?
* Can we provide world-class data centre services at competitive costs?
* Are our data centres world-class with respect to power, cooling, and environmental friendliness?
Based on discussions with customers and consultants who have worked with leading global organisations, HP has identified three critical business metrics for data centres: reducing costs; managing risks; and accelerating business growth.
These metrics reflect a long-term trend, a shift from "information technology" – which implies IT is separated from the business-to a model where technology powers the business. The role of IT management is changing from being measured in terms of delivering technology infrastructure, to being measured on the overall business outcomes derived from this infrastructure.
So, what's keeping data centre managers up at night?

Reducing cost to the business in today's high-demand environment
The challenge: Demand for storage and compute capacity is rapidly increasing as processes become digital and businesses perform more and more critical transactions and activities online. Meeting this demand has led IT departments to quickly build and expand data centres and integrate new technologies ahead of schedule. This added complexity has resulted in rising infrastructure and management costs. Power and cooling costs have also sharply increased as data centres add new energy-hungry equipment, asset utilization has become less efficient, and energy costs have risen.
The goal: More cost-effective data centres can improve your business' bottom line. For example, on average, data centre costs represent about 5% of the average IT budget. This figure includes buildings, facility management and support, power and cooling equipment costs, and electricity; it does not include IT infrastructure-related costs such as networks, servers, and storage. Moreover, new data centres easily cost $100-million to build, with some even reaching $500-million. These are investments that can not be absorbed by the typical IT budget and often need board-level approval.
The US Department of Commerce (Bureau of Economy Analysis) reports that half of all capital expenditures are IT related. Defining strategies to manage – and minimize – these costs can significantly boost profitability. Hence, enterprises should consider sourcing strategies for facilities and staff, consolidation, automation of data centre operations, virtualisation, and modernisation of IT infrastructure and facilities to achieve higher asset utilization at lower operating costs (power, cooling, administration, floor space). Applying these transformation techniques has proven to save up to 35% while providing more processing power and storage capacity.

Managing risks in a 24×7 connected environment
The challenge: Historically, IT supported internal business processes, but today it also supports external processes, including enterprise Web sites, as well as business-to-consumer, business-tobusiness, and even customer-to-customer processes. All this means that failures in the data centre can now seriously impact customer satisfaction and company reputation. Another challenge is caused by the need for more business integration, which in turn has been a driver for stronger application and infrastructure integration and has led to the adoption of modular, service-oriented architectures (SOA) to simplify development, integration, and reuse of "core" enterprise services. As a side effect of this evolution, it is not uncommon to have unexpected or large application dependencies. Hence, businesses expect IT to understand the consequences and be a partner in increasing customer satisfaction and loyalty by providing reliable service and determining the right business continuity strategy for those strongly integrated business processes or applications. The Data centre Institute recognizes these challenges and predicts that within the next five years, one out of every four data centres will experience a business disruption serious enough to affect the entire company's ability to continue business-as-usual.
The goal: A more secure and reliable data centre can dramatically improve availability and minimize interruptions to critical business processes. The objective is to evolve
from a model where security, availability, and business continuity are architected for every application or business process to a model in which the data centre is considered as a whole. This approach, along with an adaptive sourcing strategy, can improve business continuity coverage as well as customer and employee satisfaction, and it can also reduce cost.

Supporting business growth
The challenge: Today, about 95% of business processes rely on IT services. It's no surprise, then, that better IT services generally lead to better business results. For example, recent studies by the Harvard Business School and MIT Sloan School of Management have shown that market share increases were greatest in industries that used IT most extensively. In the last 10 years, the number of physical servers installed in data centres has quadrupled (IDC, CIO Strategies for the Next Generation Data Center, 2007). So it is fair to say that the data centre plays a critical role in increasing business revenue and speeding time to market. But, with many data centres being 15 years old or older, sometimes suboptimally located, often designed for legacy infrastructure, running out of power or floor space, having tons of cabling under the raised floors, unable to support high-density cooling, and with little budget for modernisation, data centre managers are increasingly concerned they may not be able to support future growth. According to the Data Center Institute, by 2010, more than half of all data centres will have to relocate to new facilities or outsource some applications.
The goal: IT executives must enable a long-term "pay-as-you-grow" strategy, using modular data centre concepts. To ensure the data centre remains a competitive advantage to the business, an "evergreen" strategy for technology refresh should be implemented, facilitated by modular design. In addition, environmental friendliness and "greening" of data centres will be necessary, in part due to new regulations (EPA, WEEE, LEED, .), to protect the corporate image or simply to provide sustainable growth. Failing to do so may lead to significant business impact, as predicted by the Data Center Institute: "Power failures and limits on power availability will interrupt data centre operations at more than 90 percent of all companies over the next five years." Combine this with the fact that more than 50% of AFCOM members have needed to add power capacity in the last 2 years, and it becomes easier to understand why power and cooling related consumption and efficiency must be addressed sooner rather than later. Not every enterprise will have the resources and know-how to deal with these issues. Hence, the importance of examining a data centre sourcing alternative as part of the overall strategy. Adopting these innovative strategies to develop a data centre that's aligned with business goals can give your business a lasting competitive advantage.

Data Center Transformation (DCT): improving four domains for lower costs, less risk, and faster growth
Data centre managers have a lot on their minds. Expanding the data centre typically means more of everything-more data and compute power, more assets to manage, more data to  back up, more power and floor space consumed, more complexity, more network traffic, and more technology upgrades, migrations, and installations. Data Center Transformation-HP's approach to helping our customers build and manage better data centres-is intended to produce a simpler, more cost-effective data centre that also delivers better services and business outcomes.
HP's Data Center Transformation strategies are based on industry-standard best practices and designed to help customers address critical challenges on the road to reducing costs, mitigating risks, and accelerating growth. Four key factors should influence your data centre transformation strategy:
* Improving energy and space efficiency
* Achieving "always on" availability
* Consolidating and virtualizing data centres and IT infrastructure
* Implementing service-centric, automated data centre operations using the best sourcing options.
In the context of the data centre, transformation activities and techniques may relate to both existing and new facilities, and include simplification, modularisation, integration, consolidation, virtualisation, standardisation, modernisation, and optimization.

Improving energy and space efficiency
Data centres located in or around some of the largest cities consume a significant amount of the total power of the city. In fact, power consumed by servers is estimated to account for about 1.2% of the world energy consumption and 1.5% of consumption in the US.
The EPA predicts that US.energy consumption by servers and data centres could nearly double again by 2011 to more than 100-billion kWh, representing a $7.4-billion annual electricity cost. As these power costs continue to increase, it is critical for data centres to actively manage power consumption and cooling.
While power and cooling related issues (insufficient power, excessive heat) account for almost 60% of significant data centre problems, more than one out of five enterprises struggles with insufficient raised floor space, according to the Data Center Institute (2006).
Data Center Transformation provides a systematic approach to keeping data centres dense and power costs as low as possible, even in very large data centres.
Reasons to transform include:
* A typical server runs at only around 15% to 20% of its computing capacity, but consumes 60% to 70% of the power of a fully utilized server. Known as over-provisioning, this practice is ubiquitous and largely uncontrolled.
* In the US, the cost of power typically exceeds the cost of equipment acquisition within two years. (In Europe, one year of power consumption typically equals
acquisition costs.)
* Environmental guidelines applicable to large-scale data centres are becoming more prevalent and more stringent in the face of global warming.
* Power, cooling, and space issues can hinder sustainable business growth.
Some of the Data Center Transformation activities a company would carry out include:
* Thermal assessments, smart/modular cooling at asset, rack, and data centre level
* Data centre assessments
* Infrastructure modernisation, performance tuning, resource and workload management, virtualisation
* Modular data centre design, including site planning, optimized racking, floor space consolidation
* Greening of data centres: environmental regulation, improve recycling and disposal, reduce hazardous materials
The benefits of Data Center Transformation include:
* Lower costs
* Smaller carbon footprint
* Improved compliance with environmental regulations
* Enabling sustainable business growth

Achieving 'always on' availability
Continuous availability is a requirement for a growing number of businesses. One major goal of Data Center Transformation is to provide end-to-end availability-and allow
seamless disaster recovery – for the entire data centre, rather than for individual applications.
IT organisations should look at proactive strategies to build tiered security, availability, and continuity solutions holistically into the data centre architecture, rather than
providing more reactive point solutions for each application. The data centre should offer a set of continuity and recovery foundation solutions, at different cost/benefit ratios, rather than having to go through an expensive and time-consuming analysis for each project or application.
Other transformation techniques include virtualisation, moving to more logical security, reducing events that are sensitive to human error – for example, by restricting the number and time of physical access to the data centre (lights-out)-and more adaptive techniques to ensure compute and storage resources for business
recovery purposes.
Reasons to transform include:
* Businesses are demanding 24×7 transactions and services.
* Growing amounts of data are exposed to the public via online interactions.
* Numerous business processes depend on real-time data.
* Potential high cost of downtime, including loss of revenue, customer trust/loyalty, business/customer data and reputation.
Some of the Data Center Transformation activities a company would carry out include:
* Business continuity planning, business impact analysis, business risk analysis
* Data centre disaster-tolerant solution/services design and implementation
* Business recovery services: outtasking or outsourcing
* Security governance, trusted infrastructure
* E-vaulting
* Business continuity rehearsal
* Data replication and continuity
* Information lifecycle management.
Benefits of Data Center Transformation include:
* Less downtime (and related costs)
* Broader coverage of applications and business processes-less exposure
* Fewer business interruptions
* Greater customer trust.

Global data centres with virtualised infrastructure
It's a simple concept – a few are easier to manage than many. A key component of Data Center Transformation is to consolidate your data centres, and for each data centre, to consolidate the infrastructure as much as possible.
In a consolidated environment, you can more effectively leverage virtualisation technology or resource pooling for more business satisfaction, easier management, better resource utilization, and dynamic capacity allocation.
The most effective approach is to look at your data centre strategy globally to determine how many data centres are needed, of which class and tier, and at what optimal locations.
Global does not imply that the target data centres should be located in one country, nor does it mean that only international enterprises can apply this approach. It only means that you should consider all the places where your enterprise is active, now or in the future, or has some form of data centre, big or small (e.g., computer room).
Reasons to transform include:
* Multiple data centres are difficult and expensive to manage and synchronize.
* Multiple data centres present multiple points of failure, each requiring significant investment to fix and continuously improve.
* It can be difficult to identify which workloads can be safely virtualized and which ones should be co-located on a single infrastructure physical device. Combining the wrong workloads together may cause availability and performance issues.
* Integrating virtualisation into existing data centre management processes.
Some Data Center Transformation activities include:
* Data centre strategy definition and master planning
* Data centre and infrastructure consolidation
* Data centre relocation and migration
* Data, infrastructure, and application migration
* WAN/LAN/MAN network design and transformation, WAN acceleration
* Data and voice convergence, unified communications
* Infrastructure virtualisation
* Data centre assessments and inventory/discovery of assets, workloads, and applications.
Some of the benefits of Data Center Transformation include:
* Easier and more cost-effective data centre management
* Better resource utilization
* Better application performance (through dynamic capacity allocation)
* Increased availability, decreased downtime (both planned and unplanned)
* A service-oriented and automated data centre with adaptive sourcing.

Implementing service-centric, automated data centre operations using the best sourcing options
A growing number of enterprises are treating IT, including the data centre, as a service. According to a 2005 Forrester report, more than 80 percent of interviewed North American firms with revenue larger than $1 billion were either in the process of, or had already adopted, a shared service model for at least one of their IT functions. A fully transformed data centre provides a transparent catalog of data centre services and may outsource major functions if it makes technical and economic sense.
How these services are sourced and charged-back to the business is an important question to resolve.
HP's experience internally and with customers has shown that it's important to balance five criteria:
* Driving the desired behavior
* Prioritizing the services that will be used by most employees or customers
* The competitive and budgetary pressures of each of the businesses served
* The market competitiveness of delivering these services in house
* Ability to set pricing, measure consumption, mediate usage, and generate (internal) bills.
Based on these criteria, it can then be determined how to source the service and how to best charge for it: allocation per employee, subscription-based chargeback or pay-per-use. Another aspect of a service-oriented IT is modularity and composability. Not only should data centre services be compliant with Service Oriented Architecture (SOA) principles, but the services should also be designed to be functionally compatible and a complement of each other, so they can rapidly be composed to support larger projects.
A final dimension of service-oriented IT is the adoption of ITIL and end-to-end service management for the delivery of the data centre services. Here the objective is to achieve "lights-out" operation based on ITIL-based processes that are standardised across the data centres. Once the service-centric data centre is in place and operational processes are standardised, it is simpler and more economically feasible to increase the level of automation of these services and their supporting processes.
Reasons to transform include:
* Businesses aren't sure whether or not to outsource specific data centre functions.
* Data centres may be dedicated to a single business unit or country.
* Cost allocations for data centre services may be arbitrary or non-transparent.
* Data centre management and operations are manual, costly, and not standardised.
* Governance of IT and data centres is fragmented and possibly redundant (shadow-IT)..
Data Center Transformation activities would include:
* Selective transformation of data centre capital (capex), operations, and management (opex): resource complementation, outtasking, outsourcing
* Optimize data centre and IT services sourcing (selectively or holistically)
* Design and implement remote management and support
* Implement, use, and continuously improve ITIL-based IT processes and best practices
* Implement end-to-end service management for data centre services
* Automate data centre services and processes
* Implement RFID-enabled asset and tape management
* Implement configuration management
* Implement data centre and IT shared services
* Transformation to a service-centric, shared-service organisation
Some of the benefits of Data Center Transformation include:
* Lower management and equipment costs
* Ability to focus on core business and key services
* More transparent and optimized chargeback model; usage-based cost allocations where appropriate
* Greater flexibility, choice, and transparency for the business
* Faster turnaround time for business requests (greater agility).

Data Center Transformation methodology
Optimizing a data centre with regard to business outcomes, key domains, people, process, and technology can rapidly become a complex exercise. Hence, a critical success factor in transforming the data centre is to have an integrated and structured approach. HP's data centre transformation methodology is based on years of internal and external experience and is well-aligned with industry standards such as ITIL version 3.
It is organized into five logical and manageable phases, each of which add value and can be broken down into projects that can typically be completed in less than 9 months.
* Strategy. This phase encompasses planning for current and future data centres and mapping out how the data centre will grow with your business.
* Design. This stage refers to building the best possible facilities and IT infrastructure for your technology and business needs.
* Transition. This stage involves shutting down redundant data centres and deploying new, moreefficient ones.
* Operations. This stage involves the ongoing maintenance and enhancement of one or more solutions.
* Continuous improvement. This phase is characterised by consistently improving service levels and responding to new business requirements.

* From an HP White Paper