Which Infrastructure Convergence Is Best for You?

Infrastructure convergence is the grouping of several IT components into one computer package. With several possible components (servers, networking equipment, data storage centers, and software), there are also several different types of convergence. Take a look at the three current options. With more information, you may be better able to determine which option is best for your situation.

Unified Architecture

This convergence method is based on the traditional rack-mounted server situation. However, with a move toward virtualization, the modern blade-and-chassis environment is markedly different. Users are now able to tie directly into fabric interconnects. Users can create their own hardware and service profiles. These powerful systems maintain their agility but come with a hefty price tag. A high-end architecture, although expensive, might be the best option for large service providers with many different resources and hundreds or thousands of racks.

Converged Infrastructure

This type of grouping is a node-based unit with both storage and computing equipment in one location, sometimes called an appliance. When consumers want to expand, they simply add another node. This offers a lot of versatility and some financial savings as big workloads can be merged into smaller infrastructure nodes. Who would benefit the most from this type of setup? Users are often mid-sized organizations looking for a way to offload VDI and move toward a price-conscious option. Of course, the converged infrastructure is easy to upgrade, so downsizing initially won’t prevent an organization from growing in the future.

Hyper-Converged Infrastructures

While the other two groupings rely heavily on the hardware, this infrastructure converges all of the components of data processing in one single computer layer. The benefits are simplified storage, easier networking, and an abstract, software-controlled process. The hardware stack can be custom built or modified without fear of damaging the infrastructure. This could lead naturally to some important financial savings. A vital part of this hyper-converged virtual appliance is the hypervisor. The hypervisor allows control of resources, API integration, and the convergence of compute, storage, and networking all in one device. Users of this system may be those organizations that are growing rapidly with the need for quick changes, usage growth, and new additions.

Focus on an Agile Infrastructure

Is it possible that hardware stacks will go the way of the floppy disk? As this hardware becomes condensed and abstracted, it may be possible. If data, VMs, and applications are able to move quickly and easily from data centers to the cloud and among many users, the future of data centers will continue to evolve. As businesses rely more and more on data centers, the drive for change is sure to remain constant.

Posted in Data Center Build, data center equipment, Data Center Infrastructure Management | Tagged , | Comments Off

The Changing Face of the Colocation Industry

In the world of colocation data center services, there has been a lot of change. Some of these emerging trends are really affecting the way that business is done and in competitive ways. The biggest players in the market are making changes and purchasing acquisitions to maintain their place in their sector of the business world. What are the biggest trends?

High-End Luxuries or Affordably Pure

As providers incorporate higher-level technologies such as cloud, hosting, and interconnection services, other providers are turning to basic services. These providers can deliver low-cost space, power, and cooling to customers looking for good-quality, affordable options.

Customer Use Flexibility

Another trend is the desire for increased capacity and density options. Customers are becoming more familiar with the role that data centers play. They want flexibility in capacity, the ability to adjust the amount of colocation capacity, and the freedom to pay for only what they are using.

New Management Software

Data center infrastructure management (DCIM) software is finding its place and impacting both colo providers and customers. For example, the providers are enjoying increased efficiency, more information for decision making, and plenty of high-tech add-ons. Customers enjoy clear visibility into the levels of power they’re consuming, one of the things that most affects their colo costs.

Increasing Numbers of Acquisitions

Mergers and acquisitions have been a normal part of the colo industry, but lately there’s been an increase in acquisitions. These exchanges can lead to more opportunities and a broader reach, both geographically and in product selection. Big acquisitions can also lead to gorilla providers in control of pricing and availability.

New Business Relations

Today representatives of the colocation providers are discussing terms and pricing with different buyers than they did in the past. The rep from the data center may be speaking with cloud architects and others within the customer organizations. This trend is affected by the natural alliance between cloud and colo services.

Edge Market Services

Another important trend is the storage of a wide variety of data from edge markets in colo centers. To provide these services, edge data centers must also provide connectivity, internal interconnection, and WAN capacity, as well as attracting content providers, long-haul carriers, and last-mile ISVs.

The Internet of Things

Both inside and outside of the IT world, the Internet of Things is a popular topic. The data necessary to run and track the use of everyday objects needs to have a home somewhere. This network connectivity allows companies and consumers to both send and receive data. It wouldn’t be possible without the colocation providers.

Posted in Data Center Infrastructure Management, data center maintenance, DCIM | Tagged , | Comments Off

Inefficient Cooling in Small Data Centers Impacts the Entire Industry

Some of the biggest costs for data centers come from overhead energy consumption, and a great deal of that energy goes toward cooling the centers. In fact, some statistics show that cooling systems take up about half of a data center’s energy intake. Really large data center operators can afford to run off of super-efficient designs, but what can smaller centers do? Inefficient cooling is a big problem for many smaller data centers.

Small Data Centers Exist in Every Community

Visit a university campus IT center or a local government IT facility. These smaller centers don’t get a lot of attention, but they are home to a large amount of the world’s IT equipment. This also means that the smaller facilities are responsible for a big chunk of the energy used up by the data center industry. It follows logically that inefficient cooling remains a problem for the entire IT industry.

The Main Problem

The problem then is that these smaller data centers are operating with inefficient cooling systems that impact the entire industry, but they don’t have the resources to significantly improve. The small-town university IT department doesn’t have room in the budget for major infrastructure upgrades. Further compounding the problem is the fact that some of these data center teams don’t even see the energy bills and remain unaware of their role in this problem. Without that awareness, there’s no motivation to make an effort to reduce energy consumption or improve cooling efficiency.

Hot Spots and Redundancy

The heart of the problem is that too many data centers are being overcooled. This happens for two reasons. The first is that hot spots must be treated, and the rest of the center is overcooled as a byproduct of that intense cooling – basically a result of improper air management systems. The second reason is redundancy. This preventative step is necessary in any data center, but with the same solution (improved air management), inefficient cooling can be reduced.

The Search for a Solution

A solution has already been determined: Simply install the proper controls and increase knowledge of the centers’ actual cooling needs. This will keep redundant units in standby mode, only kicking them on when they become necessary. Sadly, too many smaller data centers don’t have those systems or the resources to implement them. Reliability must be the top priority for data centers, but the development of efficient cooling practices must become more important, because without some improvement from these smaller data centers, the entire industry will continue to be plagued by this problem.

Posted in Data Center Construction, Data Center Design | Tagged , | Comments Off

How Flash Storage Has Changed the Face of Enterprise IT

Enterprise-class IT, or enterprise IT as it’s commonly known, refers to a combination of the hardware and software systems that have been designed to fulfill the needs of large organizations. As its name suggests, this type of IT system is used in situations that require a great deal of processing power. Those qualities that describe any type of IT system are expected to be present and fully functional but to a greater degree than found in smaller systems. For large and complex organizations, the IT requirements for performance, security, compatibility, reliability, availability, and scalability are a driving force behind productivity.


The Evolution of VDI

The virtual desktop infrastructure (VDI) that was once the norm for the biggest projects and organizations has evolved to keep up with current advancements in the technology. As networks and storage have been changed by the rapid evolution of digital business technology, VDI offerings include the use of all-flash storage. With the introduction of flash storage, the end users will enjoy much higher levels of performance.


How Does Flash Storage Fit?

To truly understand the benefits of flash storage, it’s necessary to understand what it is. Most people are probably familiar with flash drives. Basically, flash storage is any kind of storage system or data repository that runs with flash memory. The storage memory is a form of electrically erasable, programmable, read-only memory, but it is a non-volatile memory type. This is an advantage because it means that no power is required to keep the stored data intact. The flash technology is able to erase large blocks of data at once and can rewrite without completely erasing older data.


The Benefits of Flash Storage

Another advantage of flash storage is that its increased performance does not require a correlating amount of increased cost. High levels of data reduction and lower storage costs translate into savings at the data center in physical space, power, and cooling. This translates into a storage solution that is very cost efficient. You’ll find that the systems provide higher levels of data control and management. New layers of data abstraction provide deduplication, acceleration, and data encryption. Flash storage offers seamless integration with the cloud and with virtualization layers. Plus, this storage solution comes in a tiny package.


Solutions for Enterprise IT

If your large business has outgrown your old IT system and you’re ready for enterprise IT, give a thought to flash storage. This vital part of large, capable hardware and software systems and modern storage management provides organization IT departments with a new and evolved solution.


Posted in data center equipment, DCIM | Tagged , , | Comments Off

Data Centers: Securing the Cloud

truth-about-hackersIn an increasingly digital world where just about all of our personal and business-related information is stored, relayed and transacted online security concerns continue to grow and grow.  We hear about hack after hack and the need for data centers to increase their security.  As more and more location move towards cloud computing, how can they increase not only the security of their infrastructure but their overall security?  There are growing concerns that the public cloud may actually be more secure than the IT facility cloud.  Infoworld explains the concern and what the main contributing factors to the problem are, “What public clouds bring to the table are better security mechanisms and paranoia as a default, given how juicy they are as targets. The cloud providers are much better at systemic security services, such as looking out for attacks using pattern matching technology and even AI systems. This combination means they have very secure systems. It should be no surprise that the hackers move on to easier pickings: enterprise data centers. The on-premises systems that IT manages is typically a mix of technologies from different eras. The aging infrastructure is often less secure — and less securable — than the modern technology used by cloud providers simply because the old, on-premises technology was designed for an earlier era of less-sophisticated threats. The mixture of different technologies in the typical on-premises data center also opens up more gaps for hackers to exploit.”  So, does it just boil down to a narrowed focus paired with hyper-awareness of threats?  Is it just that the cloud can simply focus on its unique set of challenges whereas the traditional facilities have a wide range of weaknesses that pose potential threats and therefore security is spread thin across the board?

Cloud computing has more than proved its value so it is certainly not going anywhere.  Facilities are getting on board with it and more making the switch.  The problem is that they still have a wide range of infrastructure that must also be kept safe and protected, and traditional security approaches for facilities are different in the digital space.  What once worked for security may be so outdated that it is no longer effective and with hackers acutely aware of the gaps, like heat-seeking missiles, will swiftly find and attack those weak spots.  A breach is often the result of an un-tested system so facility managers must get more vigilant about education and testing.  Ignorance is far from bliss in this case.  The threat landscape is constantly changing so IT facilities can better protect themselves through a combination of education, real-time monitoring, protection of servers, and a dynamic multi-level approach to security.  Information must be protected within storage devices inside a facility, throughout information transmission between facility servers and clients, and throughout use within an application.  And, as mentioned above, a healthy dose of paranoia never hurt anyone when it comes to protecting secure information.  Through an extensive effort of limiting exposure on every possible front and a commitment to staying ahead of the hackers as much as possible, data center security can begin to reach the level of protection that customers expect.

Posted in data center equipment, Data Center Infrastructure Management, Data Center Security, DCIM | Tagged , | Comments Off

Data Centers Utilizing Wind Power

windenergygoogleEco-friendly and energy efficient remain the focus of data centers across the nation and around the world.  Every step a facility takes towards improvement is a step towards reduced energy consumption and significant savings.  Many facilities specifically choose to place their locations where the climate allows for natural cooling using outside air which lowers the use of air conditioning systems.  Now, many facilities are making a move toward using wind power.  These locations are using utility power derived from wind generation.  This form of renewable energy is eco-friendly because it is sustainable and dramatically reduces the need for other sources of utility power.  In some cases, data centers are becoming 100% wind powered!

There are some restrictions in place for businesses can source their wind power but this move is incredibly positive and will certainly become more and more popular over time.  Facebook has utilized wind power for one previous location and has opted to design its newest facility to be 100% wind powered because they recognize that it is an inexpensive and effective form of clean energy.  Fortune elaborates on Facebook’s latest undertaking, “Facebook announced on Tuesday that it’s building a large $1 billion data center in Ft. Worth, Texas. The facility, which is already under construction, will be Facebook’s fifth data center, and will be built on land purchased from a real estate company run by the eldest son of former Presidential candidate Ross Perot. The data center will use wind power from a large wind farm that is also under construction on 17,000 acres of land in Clay County about 90 miles from the data center. By agreeing to buy the power from the 200-megawatt wind farm, Facebook helped bring the clean power project onto the grid. A report issued earlier this month from the European Commission Joint Research Centre found that there were about 370 gigawatts of wind turbines installed by the end of 2014. One gigawatt is the equivalent to a large coal or natural gas plant… Facebook will presumably buy the wind power at a fixed low rate over several decades. If grid energy prices rise, the deal could actually save Facebook money on its energy bill.”  Additionally, Data Center Knowledge notes that it is not just IT facilities that are making this move but customers as well, “Salesforce has contracted for 40 megawatts of wind power from a West Virginia wind farm, becoming the latest cloud giant to enter into a utility-scale renewable-energy purchase agreement… The purchase covers more capacity than all of the cloud-based business software giant’s servers consume in data centers that host them.”  This shift in the industry shows that businesses, customers, and even employees are demanding more renewable energy sources for data centers and, in addition to being eco-friendly, they are significantly impacting company’s bottom lines.

Posted in Computer Room Design, Data Center Battery, Data Center Build, Data Center Construction, data center cooling, Data Center Design, Datacenter Design, Power Management, Uncategorized | Tagged , , , , , , | Comments Off

Tips to Prolong the Life of a Data Center UPS Battery

3 Data centers rely on their UPS battery to keep their infrastructure up and running.  Implementing uninterruptible power supplies with a good, reliable, long-lasting UPS battery is an expensive endeavor but one that is more than worthwhile if the power supply does what it is supposed to and provides protection.  We have discussed UPS system TCO in the past, and it is important to evaluate TCO when determining what Uninterruptible Power Supply system to implement but TCO is only accurate when you take life-extending measures to keep your Uninterruptible Power Supply system running as it should, for as long as possible.  A neglected backup power source, or one that is not properly implemented, may have a dramatically reduced life which is frustrating and costly.  It is important that a data center manager make prolonging the life of its backup power supply battery a priority so that investment is maximized and power is properly protected.  Below are some tips to prolong the life of your battery without jeopardizing the uptime of your facility,  so that you can have peace of mind that you facility is covered and you are maximizing the investment you have made.

  • Purchase the Correct UPS Battery for Your Unique Data Center
    • This is often where mistake #1 occurs. It is important to consider total cost of ownership when choosing the right backup power system and power unit for your data facility but total cost of ownership is not necessarily the full picture.  Some high-rate discharge batteries have a shorter lifespan so if a longer lifespan is a high priority it may be best to opt for a different kind of UPS battery.  A flooded or wet cell option will cost more than a VLRA battery but it will be more reliable and have a longer lifespan.  With a good picture of your data center’s specific needs, and a proper analysis of TCO you can narrow in on the proper continuous power unit to provide reliability and long lifespan within your budget.  And, once you have chosen the correct one, make sure it is installed properly.  An incorrectly installed backup power battery will often have a shorter lifespan.
  • Maintenance, Maintenance, Maintenance
    • If there is one thing that might make the biggest difference in prolonging the life of a UPS battery it is maintenance. Maintenance must be performed routinely according to a pre-determined schedule so that you are certain your backup power supply is not being neglected.  They are very sensitive to temperature and so it is important to have a monitoring system in place that alerts you if the temperature fluctuates outside of a certain range (keep it as close to  75 – 77 degrees Fahrenheit as possible).  By maintaining the correct temperature you can significantly prolong it’s life.  While automated monitoring of certain factors is important, a routine visual inspection should be part of your maintenance schedule as well because you can look for obvious damage such as loose intercell connections, damaged post seals, corrosion or fires.
  • Do Not Use Your UPS Battery Beyond Its Capacity
    • A battery is still functioning and your UPS is still doing its job, sure it may be low on life, but it is still working so why waste it, right?   It is critical that you do not push your backup power battery beyond its capacity or you greatly risk having no backup in the event of a power failure.  You should never use it beyond 80% of its rate capacity.  Once it hits 80% it will begin to deteriorate more rapidly, putting your data center at risk.  For this reason, it is imperative you not exceed a Uninterruptible Power Supply battery’s capacity.
Posted in Data Center Battery, data center equipment, Data Center Infrastructure Management, data center maintenance, DCIM, Uninterruptible Power Supply, UPS Maintenance | Tagged , , , , , , , , , | Comments Off

Data Center Cyber Attack Prevention & Protection

truth-about-hackersEverywhere we turn we seem to be hearing about another cyber attack.  Sensitive customer information, compromised.  Angry businesses.  Concerned customers.  It is a major problem and it is one that data centers must be supremely aware of and vigilant in protecting against  since facilities have access to and store so much sensitive information.  Data centers must protect networks, applications and end points from highly sophisticated, ever-evolving threats.

Protection techniques vary by each location depending on the size of the facility, the information it stores, and the specific needs of their clients.  But, in the end, all data facilities need to be actively preventing cyber attacks through a variety of means and approaches.  There are intrusion prevention systems that can be implemented by any data center that are scalable and designed to protect against the most current threats.  Data Center Dynamics explains why security must be uniquely designed for the data site, technology-driven, and innovative to truly protect data centers from potential threats, “Many Internet-edge security solutions, like next-generation firewalls, are being inappropriately positioned in the data center where the need is visibility and control over custom applications, not traditional web-based applications, and the systems that keep them operational. Security must be integrated into the data center fabric, in order to handle not only north-south (or inbound and outbound) traffic, but also east-west traffic flows between devices, or even between data centers. Security also needs to be able to dynamically handle high-volume bursts of traffic to accommodate how highly-specialized data center environments operate today. And to be practical, centralized security management is a necessity. Today’s data center environments are highly dynamic and security solutions must be as well. As they evolve from physical to virtual to next-generation SDN and ACI environments, data center administrators must be able to easily apply and maintain protections… They must also be intelligent, so that administrators can focus on providing services and building custom applications to take full advantage of the business benefits these new environments enable, without getting bogged down in administrative security tasks, or risking reduced levels of protection… Traditional data center security approaches offer limited threat awareness – especially with regards to custom data center applications and the SCADA systems that keep them running 24×7. They typically deliver limited visibility across the distributed data center environment and focus primarily on blocking at the perimeter. As a result, they fail to effectively defend against the emerging, unknown, threats that are targeting them. What’s needed is a threat-centric approach to holistically secure the data center, that includes protection before, during, and after an attack – one that understands, and can provide protection for, specialized data center traffic and the systems that keep them running. With capabilities like global intelligence, coupled with continuous visibility, analysis, and policy enforcement across the distributed data center environment, administrators can gain automation, with control, for the protection they need. Advanced attackers are infiltrating networks and moving laterally to reach the data center. Once there, the goal is to exfiltrate valuable data or cause disruption. Data center administrators need technologies that allow them to be as ‘centered’ on security as attackers are on the data center.”  Protection must be multi-level and provide protection, contingency and backup for multiple-stages of a potential attack.  While implementation of such protection may be time-consuming and costly, it is far better than having a massive cyber attack that compromises senstive customer information or that forces your data center into prolonged downtime.  Through better protection and an ever-evolving approach based on the most current information about cyber threats , customer trust is protected and business can continue to not only function but thrive.

Posted in Uncategorized | Tagged , , | Comments Off

Calculating PUE: Importance of Accurate Calculation in Data Centers

Advanced PDUData center power usage effectiveness, or PUE, is a calculation that is an essential part of determining how efficient a facility is and what improvements need to be made.  The operating costs for data facilities are constantly rising and there is a constant demand from the powers above to cut costs and lower expenses but, for data centers, this is challenging.  Data centers must continue to meet the demand of businesses, provide mission critical services, maximize uptime and protect information but with the constant pressure to cut costs.  How, with technology constantly evolving and needs constantly changing, can a facility manager assess its own efficiency and effectiveness and make adjustments to continue to improve without diminishing its ability to perform?  Data centers have been calculating their PUE in an attempt to do so for a long time but those calculations can be a bit challenging and inaccurate so – what is the best way to calculate PUE for each individual site?

When calculating PUE, a location must look at how much power is being used by servers, storage equipment, network equipment, other IT equipment, cooling and so much more.  A PUE calculation is a specific metric that can serve as benchmark for data centers and, after the first calculation, future calculations can be used to compare whether improvement is happening or not.  The trouble is, some calculations are inaccurate.  Data Center Knowledge explains how to improve calculations to ensure accuracy, ” While PUE has become the de facto metric for measuring infrastructure efficiency, data center managers must clarify three things before embarking on their measurement strategy: There must be agreement on exactly what devices constitute IT loads, what devices constitute physical infrastructure, and what devices should be excluded from the measurement. Without first clarifying these three things, it can be difficult for data center managers to ensure the accuracy of their PUE… The first part of this methodology is to establish a standard to categorize data center subsystems as either (a) IT load or (b) physical infrastructure or (c) determine whether the subsystem should be excluded in the calculation. While it’s fairly simple to designate servers and storage devices as an IT load, and to lump the UPSs and HVAC systems into physical infrastructure, there are subsystems in the data center that are harder to classify…. Some devices that consume power and are associated with a data center are shared with other uses such as a chiller plant or a UPS that also provides cooling or power to a call center or office space. Even an exact measurement of the energy use of these shared devices doesn’t directly determine the data center PUE, since only the device’s data center-associated power usage can be used in the PUE calculation. One way to handle this is to omit the shared devices from the PUE, but this approach can cause major errors, especially if the device is a major energy user like a chiller plant. A better way to measure this shared device is to estimate the fraction of losses that are associated with the data center, and then use those losses to determine the PUE… While every device in the data center that uses energy can be measured, it can be impractical, complex, or expensive to measure its energy use. Consider a power distribution unit (PDU). In a partially loaded data center, the losses in PDUs can be in excess of 10 percent of the IT load. These loss figures can significantly impact PUE, yet most data center operations omit PDU losses in PUE calculations because they can be difficult to determine when using the built-in PDU instrumentation. Fortunately, the losses in a PDU are quite deterministic and can be directly calculated from the IT load with precise accuracy if the load is known in either watts, amps or VA. In fact, this tends to be more accurate than the built-in instrumentation approach. Once the estimated PDU losses are subtracted from the UPS output metering to obtain the IT load, they can be counted as a part of the infrastructure load. This method improves the PUE calculation, as opposed to ignoring PDU.”  These guidelines will help data center managers determine a specific plan to calculate PUE.  By adhering to the pre-determined PUE calculation method, results will be more accurate across the board and over time so that progress can be seen and further improvements can be made.


Posted in Data Center Battery, Data Center Design, data center equipment, Data Center Infrastructure Management, DCIM, Power Distribution Unit | Tagged , | Comments Off

High-Density Data Center Advantages and Considerations

interxion-containment-overhAs we have previously discussed, increasing rack density and consolidating data centers is all the rage, especially going into 2016.  This is a trend we do not see going anywhere.  Many businesses are opting for colocation as a way to save money and achieve better IT management and protection.  In data facilities, space is a precious commodity.  One of the main reasons often cited for needing to relocate is simply not having enough space.  As the trend continues towards cloud storage, with the help of increasing rack density and consolidation, many data centers may just find they have more room than they think and can even implement more focused, better cooling strategies that will also help save on energy costs.  Facility rent is far from cheap so maximizing space is critical in achieving a cost-effective method of managing data.  Horizontal expansion is not the answer, vertical expansion through increased rack density and consolidation is how data centers can continue to adapt to meet their own needs without having to relocate.

Data Center Journal provides a helpful description of what high density looks like and why it makes such a big impact, “A number of different approaches to increasing power density have expanded the computing power per square foot of data center space. According to a Gartner press release (“Gartner Says More Than 50 Percent of Data Centers to Incorporate High-Density Zones by Year-End 2015”), “Traditional data centers built as recently as five years ago were designed to have a uniform energy distribution of around 2 kilowatts (kW) to 4kW per rack.” But the addition of high-density zones can increase this energy distribution several times over in certain areas of the facility. “Gartner defines a high-density zone as one where the energy needed is more than 10kW per rack for a given set of rows. A standard rack of industry-standard servers needs 30 square feet to be accommodated without supplemental cooling, and a rack that is 60 percent filled could have a power draw as high as 12kW. Any standard rack of blade servers that is more than 50 percent full will need to be in a high-density zone.”  Of course, increasing density in individual server racks, while beneficial to consolidation, brings challenges that must be addressed.  Power distribution and cooling needs are vastly different for high density racks vs. traditional server racks.  Not only must high-density power be properly supplied by energy, and properly cooled, but all of the components must have adequate backup power in the form of a sufficient UPS and UPS battery that can maintain the high-density needs should a power failure occur.  Data Center Journal elaborates on the challenges, “One constraint on power density is obviously the power-distribution infrastructure, both at the level of the utility-provided power and the backup facilities. For each watt supplied by the utility, the data center must have sufficient UPS and diesel-generator capacity to continue operations in the event of a power outage. And that, of course, is above the cabling, power-distribution units (PDUs) and so on dedicated to delivering the power to the racks. Coughlin notes that “most data centers don’t have much new power available for their facilities, so they likely have to get more power from the utility and spend a lot of money on core data center infrastructure (electrical and mechanical infrastructure, generators, power distribution and so on) just to be able to provide it. So access to more power and cost are two important variables.” But the other and perhaps more pressing need is cooling: every watt consumed by the facility is a watt of waste heat that must be removed to maintain the desired operating temperature. Herein lies what may be the biggest challenge facing higher density—particularly for facilities not originally intended to handle it. “When you increase density considerably at the rack level, much more heat is generated by the servers and a lot more cooling is required,” said Coughlin. “Cooling infrastructure is very expensive, but the biggest challenge may be trying to retrofit an old data center. Most of these older data centers were built with low ceilings and there is no easy way to improve density in many cases other than ripping up the data center—which is incredibly difficult to do, especially with live customers.”  Ultimately, if these challenges can be overcome, high-density will drive a data center’s ability to lower costs and maximize efficiency, a focus that is on the mind of every facility manager.


Posted in data center cooling, Data Center Design, data center equipment, Data Center Infrastructure Management, DCIM, Power Distribution Unit, Power Management | Tagged , , , , , | Comments Off