A Coordinated, Balanced Design in Today’s Data Center

Q&A with Mark Svenkeson, principal and partner of Hypertect (http://www.hypertect.com/):

DCP: Why is a coordinated, balanced design useful in today’s enterprise data centers?  Why should data center and IT managers care about it?  How can they benefit from it?

Svenkeson: The reliability of any item or design is determined largely by the least reliable aspect.  Your data center operations need to be supported by mechanical and electrical systems which have equivalence in their capabilities.  A well-coordinated design will ensure that all aspects of your mechanical and electrical support systems have equivalent reliability, eliminating often overlooked vulnerabilities. Most importantly, it helps protect against the human tunnel vision factor of pushing one portion of the system past the capabilities of the other, which often occurs when excess capacity is available on the electrical side.

DCP: Where should a coordinated, balanced design rank in terms of overall priority in the data center?

Svenkeson: A balanced approach to support system design and implementation should rank #1 in priority, because it is directly related to reliability.  As data centers grow and densities increase, appropriately scaling the system as a whole is critical to managing that growth effectively.

DCP: What are the biggest challenges for data center and IT managers when it comes to implementing a balanced, coordinated design?

Svenkeson: Modern support systems for critical facilities require detailed knowledge and understanding of the individual components and overall system.  The tendency may be to replace or upgrade individual components of the support system on an as-needed basis without analyzing the overall effect.

DCP: How can data center and IT managers overcome those challenges?

Svenkeson: Retain independent engineering expertise to assist in understanding the complex issues involved.

DCP: What advice can you give to IT and data center managers that have a plethora of similar solutions to choose from?

Svenkeson: Do not rely on equipment suppliers for solutions without the support of independent engineering expertise.  The solutions offered are limited by their manufacturing capabilities, and their recommendations are limited to their available product line.

Hypertect, inc. has provided independent review analysis and design of critical processing facilities since 1988; call us to discuss your concerns.

An Effective Lighting Strategy Starts at Beginning, Not at the End of Designing an Intelligent Building

- Melanie A. Reid, global alliance manager at CommScope (www.commscope.com), says:

Light-emitting diode (LED) technology is transforming both the lighting industry and the definition of an intelligent building right before our eyes. Not too long ago, LED lights were relegated to “sitting at the kids’ table” by the average person, along with Christmas trees lights, building signage, watches and perhaps the occasional disco ball. The days of being treated like one of the kids are long gone, and LEDs have matured into a power player. With increased market acceptance, innovative companies and pioneering thinkers are using LED technology as a platform to reinvent not only the light bulb, but also reevaluating how a building or data center is designed, managed and optimized starting (not ending) with an effective lighting strategy.

Our view at CommScope is that the more information you have about what is going on inside a building or data center, the better. Our friends at Redwood Systems believe the same thing and have developed an innovative fine grain sensor network (delivered as part of a lighting solution) that measures temperature, occupancy and the light levels inside a building. All the data from the Redwood network is made available to facility professionals and can be integrated with a building’s automation system to monitor, manage and optimize heating, cooling and security. Did I also mention that this system is powered, controlled and verified as an application using a CommScope Intelligent Building Infrastructure Solution platform? By integrating Redwood’s platform with CommScope’s Intelligent Building Infrastructure Solution, facility managers can now deploy a complete intelligent platform that operates many different aspects of their buildings, including LED lighting that will also assist in reducing energy cost from lighting.

This is a bold statement; however, I like to think of LED technology as similar to the invention of the iPod (I know this is a stretch, but bear with me). When the iPod was first introduced, none of us, not even Steve Jobs, could have predicted how this new device would not only change the way we consumed music, but would pave the way for how we consume all forms of media and interact intelligently with each other via integrated, smart apps. If we think in terms of how the iPod changed music and media consumption, how do you think LED technology and continued integration with an “Intelligent Building Theory” will change the way we look at designing data centers in the coming years? What other innovative solutions, using LED technology, have you seen that are transforming how you design and manage your data center or intelligent building?
 

Bio:

Melanie Reid is a member of the Channel Organization within CommScope’s Enterprise Solutions division. She has more than 10 years of telecommunications and marketing experience and currently manages CommScope’s Global Alliance Program. Melanie holds an MBA from Boise State University with an emphasis in High Technology Marketing.

Achieving Data Center – FlexPod Specialization

- Kent Christensen, practice manager at Datalink (www.datalink.com), says:

We at Datalink have recently announced that we’re the first NetApp partner to achieve Data Center – FlexPod Specialization, further cementing our expertise in migrating customers to unified, virtual data center architectures and private cloud environments.

This specialization recognizes that we’ve fulfilled one of NetApp’s most rigorous levels of competency they set for partners. Prerequisites and specializations span a number of fronts, including: Cisco Unified Computing Technology, Cisco Data Center Architecture and NetApp Professional Services Certifications in Storage Infrastructure, Server Virtualization (VMware) and Backup and Recovery.

The Data Center – FlexPod Specialization qualifies us as an expert resource to assist customers in the analysis, design, and implementation and support of next-generation data centers and virtual hosted desktops built on the Cisco and NetApp FlexPod design architecture. At Datalink, we’ve committed significant resources to developing expertise in these areas so that we can help large and mid-size organizations transform their IT infrastructure for peak efficiency, as well as cost benefits.

FlexPod is a pre-tested, optimized and pre-validated data center solution combining networking, compute and storage resources in a shared – or “unified” – infrastructure in order to better support multiple application workloads. The FlexPod infrastructure represents one of the unified architecture platforms designed, deployed, and supported by Datalink as part of our multi-phase migration strategy for large and mid-size enterprise customers.

Essentials of New Data Center Construction

- Stefan Jern, CEO of Flexenclosure (www.flexenclosure.com), says:

A WELL DEFINED PROJECT AND A TURNKEY PARTNER IS ESSENTIAL.

One of the biggest challenges for data centre and IT managers is to secure a process for the sourcing, and to whom to give the responsibility for building the data centre.

When building a data center, one may believe that somebody can do the design, somebody can build parts of the solutions and a third party may do installations. However, without deep knowledge of the complete solution, the design will not be optimal. On the contrary, there are clear risks for project delays, cost escalations and other problems.

The conclusion is that somebody with experience and knowledge of the all the components and services need to lead that kind of project for it to be successful. The project needs to be well defined, and there must be an experienced turnkey partner in charge who can handle all the phases, from design all through construction, installation, commissioning, project execution and handover.

IT and data center managers have a plethora of suppliers to choose from when sourcing a new data center. There are many players but not that many with long experience. My advice to a data centre manager would be to always check the track record of the supplier.

When it comes to the timeline, a proximal project time from order to commissioning and acceptance test on site should not be longer than 26-30 weeks. Depending on the size of the scope it could be done in as little as 20 weeks. That should be the objective.

The choice should not be made entirely on the cost on the paper, but also what the cost will be in the long term. Is the solution energy efficient and what will be my total cost of ownership? Is the design optimal for my needs and are there risks in the time plan? A pre-fabricated solution will provide a reliable solution and very short time to market while a turnkey solution from an experienced supplier will give security and low project risk.

Choose a turnkey provider that will take all aspects in to account – all the way from choosing the right energy efficient component to the design of the cooling system. A modular design is preferred. There needs to be predictability in terms of time plan, quality and customer satisfaction.

Why look for “right quality”? The answer is to mainly reduce the risks for power failure or other risk factors for sensitive equipment. The optimal choice in this regards is pre-fabricated modules that are factory built and fully equipped in a secure environment. That will both reduce the time for the data centre to be built, minimize the long term cost.

A very important aspect of a data centre and its total cost of ownership is the energy efficiency, which will have a huge impact over time. Green data centers are already a concept. However, there is more to be done in regards to developing data centers with operating expenses (OPEX) and total cost of ownership (TCO) in mind. Means for this are energy efficiency throughout the solution as well as introducing renewable energy sources also for data centers.

Finally, some advice on maintenance: To avoid operational interruptions of the switching station and to get the operations up and running as quickly as possible after a temporary breakdown, the supplier should offer proactive and reactive support.

To facilitate the problem analysis during incidents, the supplier should offer support through remote access, e.g. for the following sub systems: access system, CCTV system and control system. For the other sub-systems, such as cooling system, fire fighting system, UPS and AVR, there should be telephone support.

Some sub-systems need maintenance more than once a year; one example is the cooling system. The supplier often assists with this in close cooperation with the client, since the customer normally has some own sort of facility management, either through an internal department or through an external partner.

Replacement of vital parts of the equipment is necessary to remain in operation. In this respect, you should expect from your supplier that they will advise you how to procure the right amount of the vital parts that you need, with instructions for replacement of vital parts that could be troublesome to get correctly installed.

By Stefan Jern, CEO, Flexenclosure (www.flexenclosure.com), a Swedish supplier of green power management solutions and pre-fabricated modular data and energy centres.

The Future of Data Center Design

Interview with Jim Smith, CTO of Digital Realty (http://www.digitalrealty.com/):

DCP: Jim, you’ve spoken about modular datacenters and other next-generation datacenter designs at recent conferences, and that’s a topic that a lot of our readers ask about. Is ‘modular’ just a buzzword, or is it something that is going to have a real impact?

Smith: I definitely think it’s real because it has significant measurable benefits in terms of construction costs, reliability, energy efficiency, operating costs and more. Modular isn’t a passing fad. In fact, I feel strongly that it will become a cornerstone principle for how datacenters will be designed and built going forward.

With that said, I should clarify something. Modular datacenters, to me, are not the same as “containerized datacenters” or “pre-fabricated datacenters”. Some people use those two terms interchangeably with modular, but to me they are very different. Containerized or pre-fabricated datacenters are very cool pieces of modern datacenter design, and they have very interesting applications. But those solutions are not what I mean when I talk about modular. When I say “modular” I mean brick-and-mortar datacenter buildings that have all the benefits of a solid, permanent structure….but built with pre-constructed components that translate into higher reliability, lower costs and dramatically faster completion timelines.

There is a quiet revolution happening in the way datacenter buildings are being built, and modularity is at the heart of that. The completed buildings may look similar on the outside and the raised floor space may look just like finished datacenter space always looks, but the way you got there is radically different and a huge step forward.

DCP: Can you talk more about pre-construction? What kinds of components are pre-built?

Smith: Pre-building is a big part of taking a modular approach to building datacenters. And you’re right, we’re not pre-building entire datacenters, driving them down the highway on a wide-load trailer and plucking them down on a cement pad. We call our new datacenter design and construction process POD Architecture 2.0, and it is the logical evolution of the modular construction approach we pioneered with our original POD Architecture.

We have selected key components of the datacenter to pre-build offsite in a controlled, clean, highly-efficient factory setting and then deliver them exactly when they are needed in the construction timeline. Electrical rooms and mechanical/cooling systems are the most important components we are pre-building and inventorying, and we chose those for a number of reasons.

The primary reason why pre-building is better is how much it speeds up the process. Any of your readers who has been involved in a datacenter build knows that certain steps in the workflow are liable to be bottlenecks that slow down the entire project. On-site construction of the electrical rooms and cooling systems are two of those potential bottlenecks because of their complexity and how they sit in the critical path of other steps in the construction process. By pre-building these components off-site and delivering them at pre-determined points in the construction process, you keep the timeline on track and avoid costly, frustrating delays. Plus there is a higher level of quality control in a manufacturing environment versus a construction site, which results in a higher quality, more reliable facility.

DCP: When you hear about datacenter builds going millions over budget and weeks or months over schedule, are these common culprits?

Smith: Construction projects in general are susceptible to cost overruns and delays, as everyone knows. That’s the nature of construction. But datacenter projects are particularly vulnerable because of the highly technical nature of the project and the huge amount of capital involved.

We’ve all heard horror stories about projects that have gone awry. Sometimes it’s just poor planning. Sometimes it’s that the company keeps changing its mind about what it wants. Sometimes it’s bottlenecks caused by those complex steps in the build. Those problems have existed for as long as folks have been building datacenters. Through POD Architecture 2.0, we are addressing those long-standing datacenter challenges in ways that reduce costs, reduce risks, accelerate construction timelines, improve quality and achieve higher energy efficiency.

DCP: In one of your presentations you talked about how much time a modular approach shaves off the construction timeline for a datacenter. The numbers are almost hard to believe.

Smith: The difference is dramatic, and it is pretty stunning when you first hear it for the first time from industry folks like me who are sharing case studies. But when you walk through it, it makes perfect sense. Using POD Architecture 2.0, we can deliver a datacenter in as little as 12 weeks. The industry average using traditional design and construction methods is in the 9-18 month range. That’s not only a fraction of the time, but millions of dollars or tens of millions of dollars in cost savings.

That stuns a lot of people when they first hear it, but everyone who has been involved in a datacenter build knows the most painful steps and the causes of the biggest delays. They know how many weeks can be lost on bottlenecks. They know the problems and common delays with traditional construction processes. And this next-generation design and construction methodology simply goes after each of those problems and re-thinks them in a way to make the entire process simpler, faster and less expensive. When you start removing those most common delays from the equation, the math takes care of itself and these dramatically shorter project timelines make perfect sense.

It appeals to me because it is highly conservative from an engineering point of view. This isn’t radical design. We are simply catching up to what other industries have known for decades: that quality goes up, costs go down and product time drops significantly when you utilize modular best practices.

News: Data Centers Canada Expands its Toronto North Data Center Facility

- Tony Di Benedetto, managing partner at Data Centers Canada (www.datacenterscanada.com), says:

Data Centers Canada, a Canadian based data center owner / operator has announced that it has commenced construction to expand its Toronto North Data Center Facility. Data Centers Canada Toronto North Facility is located within the Greater Toronto Area in the City of Vaughan.

The Toronto North Facility will expand to 20,000 square feet commented Tony Di Benedetto, Managing Partner at Data Centers Canada. The Phase 4 POD expansion will add additional capacity to house a mixture of private cabinet and cage environments. The Phase 4 POD area will accommodate up to 300 watts per square foot power density.

“We are very pleased with our continued expansion of our Toronto North Facility.”, noted Di Benedetto. “We have worked very hard in creating a scalable, flexible solution for enterprises who seek to expand their data center footprint in the Canadian Geography over the past several years.”

Diverse Meet-Me-Room’s within the facility allow tenants access to both leading Canadian and global fiber networks. Data Centers Canada offers its tenants access to both its multi-homed BGP network access platform as well as access to third party carriers via a carrier neutral colocation solution.

“Our carrier partners have worked diligently with our engineering group to enhance connectivity options for our clients.”, commented Di Benedetto. “Tenants at Toronto North can easily and cost effectively interconnect with a wide range of both Canadian and global fiber networks to address their business requirements.”, added Di Benedetto.

The Toronto North Facility utilizes a variety of green technologies to enhance its power and cooling infrastructures. Data Centers Canada is focused on its green data center program in collaboration with PowerStream the second largest municipally-owned electricity distribution company in Ontario. Green technologies such as; free air cooling and air containment / air distribution systems will be built into the data center environment.

It is anticipated that the Phase 4 POD construction will be completed by March 2012.

Data Center Customers Know What They Want

- Ilissa Miller, CEO, iMPR at FiberMedia Group (www.fibermedia.net), says:

Data center customers are smarter than they used to be. Just a few years ago, with fewer facilities to choose from, data center providers offered the basic services – space, power, connectivity. Today the options are exponentially greater as evident by the November 29th greater New York data center conference being held in New York City. Organized in less than two months, the event expects to host more than 250 senior-level commercial real estate and technology service providers primarily from the greater New York area.

One such company, FiberMedia Group, is a host to five data center facilities in and around the New York Metro Area. The company’s CEO, Michael Bucheit is a featured participant on the day’s final panel taking place at 12:25pm titled: “Innovative Data Center Design and its Impact on Marketability, Management and Operations.” Other speakers include Arthur Pearson of Con Edison Commercial & Industrial Energy Efficiency Program, Sean P. Brady of Cushman & Wakefield, Michael Bucheit of FiberMedia Group, Dennis Julian of Integrated Design Group, Josh Rabina of Sentinel Data Centers and Brian Schafer of Highland Associates. FiberMedia is focused on improving the customer’s experience by providing a full array of solutions to help them grow. Within its data centers, FiberMedia provides a full array of managed services, network accessibility, cloud networking solutions as well as business continuity and disaster recovery services to ensure businesses remain operational in the most dire circumstances.

The panelists are expected to discuss and analyze what makes an ‘asset’ marketable today from the perspective of repositioning traditional commercial real estate property to data centers, the importance of identifying new and unique fiber and other related data-infrastructure requirements to the cooling systems, electrical and power backup required to sustain the technology housed in the facilities.

The summit concludes at 1:30pm. For more information about the event visit http://apartmentsummit.com/nydata/.

Where the Cloud Really Lives

- Paddy Moogan, SEO Consultant with Distilled (www.distilled.net), says:

Imagine the cloud, what do you think of? One of the hottest buzzwords of recent years provokes a wide range of images – servers, computers, the sky and clouds, but how many of us actually know what it is? In reality, the cloud is anything but cloud-like. Interxion, a European data centre company, has given us a peek inside its own data centres to give a better idea of what the cloud actually looks like.

From the Outside: Data Centers

Let’s start from the outside with a view of an Interxion data centre in the Netherlands. The exterior of a data centre is unique only in the equipment required to provide the cooling for the equipment housed in the data centre, as illustrated below.

These are chiller units on the data centre roof

If you imagine a data centre to be a huge building that sticks out like a sore thumb, think again. They can sometimes be located right where you wouldn’t expect, take the Interxion facility in London as an example:

Right in the heart of London, cooling fans on top of their London data centre

Those of you familiar with the London skyline will recognise the close proximity of the Gherkin building, demonstrating the proximity of the London financial district. Believe it or not, the close proximity of this data centre means a better service to Interxion’s finance customers. Many of these customers work on sub-millisecond trading where a fraction of a second can make a real difference to their bottom line.

Stepping Inside: Data Centre Security

Moving inside, as you’d imagine, security is tight. You can’t just stroll into an Interxion data centre. There are many security measures in place (some of which we can’t talk about), including CCTV, biometric scanners and mantraps:

Secure mantraps meet you as you enter the data centre

Where Does All the Power Come From?

Suffice to say, power is critical to a data centre. Without sufficient power, a data centre cannot function, and Interxion understand this by guaranteeing 99.999% uptime:

Interxion ECSC Data Centre Support

The sheer power and scale of a data centre is pretty hard to comprehend. We can’t go into exact details, unfortunately, but the following images may help give some context to the amount of power they require:

Backup diesel generator in a data centre

The backup generator above is key to the smooth running of the data centre. Should power be lost – in a storm, for example – it is the job of this backup generator to kick in and keep the centre running. Fortunately, this backup generator hasn’t been needed yet. But with a large number of Interxion’s London customers being in the finance sector, a short period of downtime could translate into losses in the millions. But don’t worry, the generator above is there to be a safe fallback, along with five others – just in case.

Power from the outside world is channelled into the data centre via a transformer switch room, which is made up of busbars, switchboards and breakers:

Transformer Switch Powering a Data Centre

Connecting Everything Together

As with power and security, connectivity is essential to a data centre:

Backbone Network Cables

In the Interxion data centre in London, servers are located together within racks. The image below shows yellow strips that contain cables which connect to the outside world:

The cables are kept in tidy rows so they do not impede circulation

Looking at the bigger picture, there are hundreds of server racks within any one room of a data centre. Some customers will wish to connect directly to each other to help with faster connectivity; some will also have their own private rooms containing just their own cabinets and servers.

The picture below shows us the configuration of a typical aisle:

Aisles of Servers

Hopefully these photos have given you a good insight into where the cloud really lives, as well as the scale at which an Interxion data centre operates.

To see more, take a look at the full image gallery on the Interxion website.

How LEED Certification Goals are Impacting Data Center Site Selection

- Todd Boucher, rrincipal with Leading Edge Design Group (http://www.ledesigngroup.com/), says:

Over the past year we have seen over a 30% increase in the number of customers expressing interest in obtaining a LEED Certification for their new data center build out or renovation. This change is illustrative of the continued integration of energy efficiency and environmental sustainability into data center design requirements, but moreover presents a new challenge for owners undertaking the site selection process.

While traditional priorities in a site selection process – power redundancy, natural disaster protection, cheap energy costs, etc – will never change, the impact of sustainability has been steadily increasing. For example, organizations like The Green Grid publish ‘free cooling’ maps, helping owners understand which geographic regions will allow them to maximize economizer usage throughout the year. In addition, organizations that have set LEED certification goals must take into consideration how proposed sites will impact their overall LEED scorecard.

Under the LEED 2009 Commercial Interiors rating system, a project must successfully obtain a minimum of 50 points (out of a total 110 total available points) to achieve a LEED Silver designation. While a target of less than 50% of the total available points may seem marginal, anyone who has completed the LEED process know and understands how challenging it can be to earn 50 points, how every point becomes critical, and why site selection is an integral part of achieving LEED certifications.

Site Selection accounts for 21 total available points on a LEED project. If you have identified LEED certification as a project goal for your data center, it is highly recommended that you work with your design team to evaluate the proposed site(s) against the LEED Site Selection requirements. While evaluating your proposed site(s), here are some things to think about in the context of the LEED Site Selection Requirements and the implications it may have on your lease negotiations:

SSc1 – Site Selection (1-5 Points) - In this you have the option of selecting to place your data center in a building that is already LEED certified, in which case you will achieve all 5 points. If the building does not meet this criteria, you will need to obtain points through items like stormwater design & management, green roofs or roofing with high solar reflectance index, onsite renewable energy, and water efficient landscaping. Thus, if you are leasing space and have LEED goals, it will be important to identify if the building owner will accommodate any site changes required to obtain some points under SSc1.

SSc2 – Development Density and Community Connectivity (6 points) – This credit is important from a site selection perspective because it is impacted by the community that surrounds your site. The intent is explicit – “to channel development into urban areas with existing infrastructure, protect greenfields, and preserve habitat and natural resources.” If your data center site is in an urban area, you may meet the density requirements (60,000 sqft per net acre).

If not, your site will need to meet the following three “community connectivity” requirements:

1. Be located within 1/2-mile of a residential area or neighborhood with an average density of 10 units per acre net
2. Is within 1/2-mile of at least 10 basic services (supermarket, place of worship, bank, etc)
3. Has pedestrian access between the building and the services.

You are required to submit scaled maps and pictures outlying the location of services, residential areas, and pedestrian access, so it is critical to measure the ½ mile requirement to verify you are compliant.

SSc3 – Alternative Transportation (10 total points) - this credit is divided into three separate subcredits that all revolve around enabling building occupants to find means of alternative transportation to your site. For example, 6 points are earned if you are within a ½ mile walking distance of a rail station. Two points are available if you provide the required amount of bicycle storage and shower/changing facilities. Two additional points are available if you provide the minimum amount of preferred parking for van/carpools.

Achieving LEED certification is a challenging but worthwhile achievement for a data center project. It is clear that the location of your data center (or the building housing your data center) site will have a significant impact on your ability to meet LEED certification goals. If you have set LEED certification as a goal, it is highly recommended that you work with your design team to fill out a preliminary LEED scorecard. This will provide you with a perspective of your potential certification.

In addition, it will allow you to identify areas which may require negotiation with your landlord, as they will be an important stakeholder in the process and their participation is critical. This is all work that needs to be completed up front (ideally before a final site is selected) to ensure that you can properly identify an achieve LEED certification and successfully negotiate those requirements into the design and lease negotiation process.

* Note - All statements made in regard to the LEED rating system are specific to the 2009 rating system.

New Boston Data Center Provides Uninterrupted Service

- Rich deFay, project manager for the Copper Development Association (http://www.copper.org), says:

Miles of copper ensure power reliability and power quality.

The largest and arguably the most significant data center facility in the entire New England region is located inside the 70-year-old Macy’s building at One Summer Street in Boston’s “Downtown Crossing.”

The 800,000-square-foot data center supplies high-quality electric power to an extensive customer base operating in Beantown, 24 hours a day, seven days a week. It includes eight utility feeds from two substations and service connections to no less than 40 national and international communication carriers.

“Boston’s new data center is truly a power hub for the community,” said David Brender, National Program Manager for the Copper Development Association (CDA). “The businesses tapping into the power source depend on the electrical infrastructure being stable, maintaining the correct voltage and staying online without interruption.”

The stability of the data center depends on the copper inside. The power density exceeds 400 watts per square foot, which is nearly quadruple the power density of typical units built 10 to 15 years ago. The robust copper wiring and grounding practices enable exceptional power reliability and power quality. Copper is also durable and so offers years of reliable performance. It won’t lose the voltage levels required to run equipment efficiently.

The Boston data center has never experienced a utility outage but is fully protected should one occur. Data centers currently represent about five percent of the total electrical usage in the country. That number is expected to grow 15 to 20 percent in the next five to ten years.

For more information about the Boston Data Center project, please visit: www.copper.org/applications/electrical/pq/casestudy/a6134/a6134.html.

What's Your Data Center Strategy? Build or Retrofit?

- Bill Kosik, Principal Data Center Energy Technologist at Hewlett-Packard (www.hp.com), says:

Why are the HP Critical Facilities Implementation service and Flexible Data Center useful in today's enterprise data centers? Why should data center and IT managers care about it? How can they benefit from it?

Aging data centers are forcing many organizations into a corner. Many are running out of space, have insufficient power or operate in environmental conditions that threaten business continuity.

Clients considering their data center strategy might be unsure whether to retrofit their existing data centers or to build new from the ground-up. The decision to build or retrofit is not a simple choice and will have major financial and operational repercussions. Organizations expect data centers to last a long time, so careful evaluation of the entire IT space and the company’s critical assets portfolio over the long term is required.

There are essentially three options available:

1. Implement small, tactical operational improvements
2. Build a new data center
3. Retrofit an existing data center

Segments of the data center industry are undergoing a rapid and significant transformation that can potentially meet the need for more modular, scalable, reliable, energy efficient, and cost effective facilities. Now-a-days monolithic data centers are only being used for certain, specialized uses. Other than these, the rest of the industry has shifted towards modularity and scalability.

Whatever the chosen strategy, it is important to design new or refreshed sites accordingly, delivering faster time to economic value, increased utilization and capacity as well as lower energy, operation and maintenance costs. Such data centers will have a converged infrastructure – more energy efficient, more reliable, and with greater flexibility.

HP is an expert integrator for the planning, design, building and commissioning of data centers – from concept to commissioning, and beyond. We help businesses get the most of their data centers by combining our facilities consulting, design and assurance services together with turnkey construction project management.

HP’s Critical Facilities Implementation (CFI) service simplifies the process of designing and building data centers by offering design, construction and project management from a single vendor. With HP CFI, clients can lower their cost of ownership by having a single integrator that delivers all elements of a data center design-build project from start to finish. For organizations looking to build new data centers or expand existing facilities, CFI provides a holistic approach, streamlining the process for faster builds at lower costs.

In addition to HP’s consulting and support services to design and build data centers, HP also offers a new data center design that cuts the capital investment required for the data centers in half while reducing an organization’s carbon footprint. HP Flexible Data Center (HP Flexible DC), a component of HP CFI, offers a industrialized, modular approach to designing and building data centers that allows clients to replace traditional data center designs with a flexible solution that can be expanded as needed while conserving resources.

The Flexible DC has a unique design concept termed the “butterfly” which features four prefabricated data center modules that stem off a central spine housing administrative and other functions. Several different options for power and cooling topologies are available, based on the functional requirements and the climate in which the data center is located. The final product looks like a traditional brick-and-mortar facility but is designed and constructed using a very efficient industrial process, where major portions of the data center are built and tested off-site and assembled in the field. This allows for a very fast design and construction process that optimizes both first cost and on-going costs.

Where should the HP Critical Facilities Implementation service and Flexible Data Center rank in terms of overall priority in the data center?

We believe these services should be very high priorities. The HP Critical Facilities Implementation service is a natural first step for organizations looking to develop a roadmap for data center growth. This service converges technology with strategic facilities planning to cover all aspects of client needs, whether they are updating an existing data center or building new from the ground up.

Clients, such as financial service providers, government entities, and cloud and colocation hosts, will find the scalable and modular nature of HP Flexible DC a compelling option. We help organizations innovate the way they build and operate a greenfield data center for greater savings over its life span. Clients rely on our strategic facilities insight and experience globally – this includes mature geographies as well as major growth markets like China and India, where we are currently providing these CFI services for large-scale data centers.

What are the biggest challenges for data center and IT managers when it comes to HP Critical Facilities Implementation services and Flexible Data Center?

Data center facilities are undergoing a sea-change. Today’s power and thermal requirements for IT equipment are strongly influencing and in some cases changing the way electrical and cooling systems are designed. Monolithic data centers, once a hallmark for reliability and availability, are no longer economically viable, nor sufficiently flexible for today’s business environments.

Modular design, as it is applied to a traditional brick-and-mortar data center, will offer improved flexibility and lower construction costs over a monolithic data center. But using this approach still requires both infrastructure and IT space for future expansion. While a definite improvement, modular design might still be too costly and for some enterprises.
These and many other complex decisions must be made before starting the design and construction process. Obtaining the data necessary to make these decisions requires a partner with highly specialized engineering expertise and a rigorous process to help clients understand the key technology trade-offs and financial and performance impacts.

How can data center and IT managers overcome those challenges?

HP Critical Facilities Implementation services help organizations overcome the headaches caused by a fragmented design-build process. Funneling the entire process through a single integrator creates continuity and gives organizations an end-to-end experience. Leveraging the Flexible DC design also eliminates future hassles when expanding data center facilities as an organization’s technology needs increase. The flexible design facilitates expansion and significantly reduces build time.

As a result of using HP’s CFI service and Flexible DC design, organizations are able to experience faster build times for data center facilities at reduced costs. While cutting capital investment, the faster build time also allows organizations to realize benefits quicker and expand without business disruption.

What advice can you give to IT and data center managers that have a plethora of similar solutions to choose from?

HP is the only vendor that can offer a full turnkey design-build offering without having to partner with outside consulting design or construction management firms for elements of the solution, such as design engineering and the creation of blueprint drawings. Similarly, no other vendor can execute a comprehensive data center implementation plan. With HP, organizations can avoid a fragmented approach to their data center design-builds. Working with HP provides clients with a controlled, end-to-end experience resulting in faster design and build time as well as reduced costs.

The Virtual Facility Concept: Useful in Today's Data Center

- Hassan Moezzi, CEO of Future Facilities (www.futurefacilities.com), says:

Why is the Virtual Facility concept useful in today's enterprise data centers? Why should data center and IT managers care about it? How can they benefit from it?

A data center is built with fixed available capacity which includes provision for Space, Power and Cooling. However, owner/operators rarely, if ever, achieve anywhere near full utilization of the available capacity. The cause can be directly associated with the dynamic nature of the facility; IT assets as well as the type of equipment housed are in a constant state of flux.

In practice, once a data center is commissioned, IT management take over the operations. The Facility Manager (FM) has to react to IT Management (IT) requirements as and when new load capacity plans are decided - usually on short notice. Unfortunately, little thought is given to the fact that any operational decision, such as equipment deployment, has a direct engineering impact on the entire data center.

Air is the main vehicle for cooling IT equipment, and the transient nature of IT equipment deployment affects both airflow and temperature distribution. Poor airflow management can be directly caused by changes to IT equipment undertaken without full consideration. Dynamic asymmetric load distributions can therefore cause major problems over time as the data center starts to fill up; localized hotspots start to develop long before the facility is even close to full capacity, endangering data center health and encouraging facility managers to limit further equipment deployment.

This is a holistic problem and the owner/operator is dealing with a moving target in his bid to utilize the maximum available capacity. Increased thermal risk and costs due to stranded capacity and inefficient use of power are inevitable.

To address this complex interaction between the Facilities Management and IT requirements, Future Facilities Ltd (FFL) is advocating the use of simulation techniques to predict the cause and effect of any changes within the facility before actual implementation. Predictive analysis is the only way to avoid extra cost or potential disaster in a data center.

To this end, we at FFL are pioneering the unique concept of the Virtual Facility (VF) to help lead the design as well as operation and management of any facility. The VF is the full 3 dimensional mathematical representation of the physical/logical data center at any point in time, be it past, present or more importantly the future.

To create the VF for a data center, Future Facilities have developed the 6SigmaDC software suite, tailored to fit the lifeline of the facility. 6SigmaDC employs complex Computational Fluid Dynamics (CFD) techniques to simulate airflow movement, and also provides the means to build a perfect virtual replica of any facility, as well as its logical and physical behavior. Once the VF is built and calibrated, both the Facilities Management and IT groups can communicate through this one singular hub.

Where should the Virtual Facility (VF) rank in terms of overall priority in the data center?

The Virtual Facility is an umbrella that covers the lifeline of a data center in its entirety. It is a predictive methodology as well as a visual communication tool that should be used by the FM and IT management at all times from design, construction, and commissioning across to on-going operation of the facility. The data center industry is waking up to the importance of communication within operations, and DCIM is becoming popular. The VF is the perfect central communications hub that connects operations with engineering, and can contain as much detailed data center information (existing and predicted) as is necessary to operate the facility with improved success.

What are the biggest challenges for data center and IT managers when it comes to the Virtual Facility (VF)?

From a technical point of view the VF of the data center has to be built and calibrated in the first instance, which requires some commitment. However the most fundamental problem is for the IT and FM community to accept the concept of using a singular predictive simulation techniques as the pre-requisite for any changes within the data center, and as the entire solution for both engineering experimentation and facility management. Overcoming this conceptual hurdle reaps great rewards, as the VF acts as a buffer zone to test the cause and effect of any changes prior to the final implementation. This means optimizing facility operations without putting the data center at any risk.

How can data center and IT managers overcome those challenges?

The task of building and calibrating the VF is often taken on by qualified consultants using the 6SigmaDC toolset.

Once the VF is ready, IT and FM departments can use it as the hub for further operational matters. Future Facilities engineers are available to provide support if necessary, and those wanting to take the Virtual Facility even further can take a 6SigmaDC training course. Although it requires an initial commitment, the resulting benefits are typically more than enough to provide an understanding of the full value of the Virtual Facility methodology.

What advice can you give to IT and data center managers that have a plethora of similar solutions to choose from?

In general, the results of simulation solutions will only ever be as accurate as the tools used and the information put in; therefore it is important when considering a simulation tool to avoid short cuts and 'quick fixes' - after all, it will be used to improve the health of your data center, which is priceless.

Future Facilities believes fully that accuracy is the key to successful CFD simulation. We have therefore made sure our software takes into account as much physical/engineering information as possible to ensure simulation results match reality. We also actively encourage calibration of the simulation model, for the same reasons. The CFD calculations that take place behind the scenes of our 6SigmaDC software are highly advanced, providing precision and integrity in results and hopefully providing the user with peace of mind.

Whilst the use of CFD techniques is not a new concept for the data center community, the idea of building a complete 3 dimensional Virtual Facility that includes details such as the assets, power network distribution, cable management as well as dynamic power feed, live telemetry, and much much more, is absolutely unique.

Future Facilities Limited is a unique company in that:

• We are dedicated to solving problems specifically for the data center market.
• We develop the 6SigmaDC suite of products to bridge the communication gap between the IT and FM departments for this market. To this end we work with the full spectrum of the vendor community supplying parts and services to the owner/operators. This includes IT and infrastructural equipment manufacturers, various consultancies providing architectural, mechanical and electrical services, as well as owner/operators.
• FFL has a consultancy engineering arm that uses our 6SigmaDC software in real engineering projects. This team also provides full calibration services for the client, and full software support. This engineering group tests and validates our CFD engine in real data centers, ensuring quality and accuracy. No other CFD vendor provides this complementary service to its software development in this market.

Our overall aim is for the common problems associated with facility design, operation and management to become a thing of the past, as 6SigmaDC and the Virtual Facility opens the door to improved communication, knowledge and understanding of the workings of each and every data center.

What constitutes a good data center design?

- Mark Svenkeson, Professional Electrical Engineer and President of Hypertect (www.hypertect.com), says:




What constitutes a good data center design?

One which efficiently meets the needs of the client. It should contain the following aspects:

• Be flexible to adapt to changing requirements and technology
• Be reliable in supporting customer processing
• Be maintainable, so as to retain design reliability over time
• Include vendor-neutral solutions to encourage competitive procurement
• Be well-documented and understood by owner/customer

Why is a data center's design important for an enterprise's success?

A data center is a highly complex electrical and mechanical environment. The processing operation housed therein is central to the company’s financial success; outages cost money and clientele. Processing technologies continue to evolve at fast rates, and a facility which can adapt with minimal disruption will present minimal overall costs. A facility which continues to operate seamlessly during times of utility outages and equipment failures will maximize profits.

What key considerations go into designing a data center?

Obtaining the best answers to the following questions:

• What level of reliability is required?
• What are the power density expectations?
• Does the system include maintainability of all components which contribute to reliability?
• Does the design take advantage of natural physics and environment?
• Does the location provide access to required communication and power infrastructure and have space for heat rejection and power generation?

What pitfalls can designers and builders typically fall into?

Applying vendor solutions without understanding the implications. Forgetting that, during a utility outage, when generators are starting, the cooling system may not be functioning, but the heat load is always present. Designing specific solutions for immediate concerns, while neglecting flexibility considerations required to support future needs and technologies.

What advice can you give to a company that is looking at several different design/building companies?

Consider:

• Does the company have exclusive vendor relationships? (These can impact procurement decisions.)
• Does the company provide in-house engineering and project management, or does it rely on contractor design/build?
• Dual-corded processing requires dual-path support infrastructure for complete implementation.
• High power densities require uninterruptible cooling.
• KISS is a valid design principle.

What Constitutes a Good Data Center Design?

- Chris Loeffler, Data Center Applications Manager, Distributed Power Solutions, Eaton Corporation (www.eaton.com), says:

What constitutes a good data center design?

Today’s best data centers are designed from the concept stage through the entire service life of the facility, with the following three objectives:

1. Availability and uptime of the enterprise-related applications that “pay the bills”. Make no mistake: Reliability and availability trump every other consideration in data center design.
2. Flexibility for future growth. The data center must be able to support the business’ expected growth, usually for a period of 20 to 40 years. That means infrastructure decisions must be made with great forethought and an eye toward both current conditions and possible conditions a couple of decades from now. Also, equipment and support services will likely be “changed out” several times over the data center’s lifecycle, and this must usually be accomplished without disrupting the data center’s work. Therefore, you must build in the flexibility to make those kinds of change while still maintaining services to your users.
3. Efficiency in power and water usage, and deployment of environmentally-sustainable materials and processes. LEED compliance, at some level, is a requirement for virtually all new data centers and for most data center upgrades. No longer can the enterprise simply assume that uptime is the only thing that matters. Efficient and responsible use of resources is good for the financial bottom line, and necessary for businesses to maintain their good standing in the community.

Why is a data center's design important for an enterprise's success?

The huge cost of installing new, and upgrading existing, data centers make the stakes very high. Almost all businesses nowadays rely heavily on data processing and Internet communications. They’re no longer a luxury; they’re a requirement of doing business. Many companies are rendered instantly helpless when a data center suffers a failure. There is often a huge loss in revenue and goodwill when these disasters occur. Bad press can be particularly damaging and result in customers departing, possibly never to return. So the data center must be designed for maximum reliability and extensive flexibility. It must be capable of withstanding multiple system and infrastructure failures while continuing to service customers. Downtime is no longer an option.

What key considerations go into designing a data center?

1. How large will the facility need to be? Do I want one huge building, or does that place “all my eggs in only one basket”? Do I have enough room to meet my needs over the next 20 or 30 years? How big is too big? Can I operate efficiently in the early years of the data center’s life?
2. Where do I locate the data center? Is adequate and upgradeable power and water available? Is it wise to locate the center in a downtown area where power disturbances may be frequent? Is a rural location better? Can my staff access the facility in inclement weather? How does my location affect my cost of power and cooling?
3. Don’t forget security. Can the facility and its staff be protected from intrusion and disruption? What is the backup plan when problems or failures occur? Is there a Plan B and C? Do I invest in one or more disaster recovery facilities to fall back on during a hurricane, earthquake, or other calamity?

What pitfalls can designers and builders typically fall into?

The biggest pitfall is not being able to predict the future. No one can do that, of course, but experienced engineers and consultants can provide invaluable advice, based on their real-world knowledge. Don’t forget to set up a Plan B and Plan C to address unexpectedly rapid growth, or the need for disaster recovery provisions.

In addition, avoid the temptation to invest in technologies and processes that do not show a clear return on investment for your business. Very expensive projects are too often undertaken without a clear view of when (or if) they will pay for themselves. Remember that even though a data center is a nerd’s paradise, it’s still a functioning business, and the financials must make sense. Every part of the data center must have a defined goal and a defined ROI. The data center manager will typically be measured on both meeting the company’s performance and reliability goals and on the project’s time to payback.

What advice can you give to a company that is looking at several different design/building companies?

Give at least three companies a chance to present. Ask every question that comes to mind, even the obvious ones, and look for clear answers based on the firm’s real-world experience. This is not the time to take a chance on a new designer, engineering firm or architect, so don’t risk your multi-million dollar data center on anyone who hasn’t proven themselves. Question them specifically on how their solution is more reliable, more flexible, and more efficient than the alternatives. Don’t accept vague answers. They should be able to provide tangible evidence of how their recommendations will benefit you. Look for something you can hold in your hand, or better, place in your wallet. Don’t put your faith in promises. Demand to see results.