The recently established Data Center Optimization Initiative (DCOI) is an important mandate for federal data centers that encourages the sharing of information to encourage optimization of infrastructure and reduce inefficiency in data centers. Nothing is more of a hot topic in data centers than the need to improve efficiency on all levels to remain sustainable and effective. The White House describes the requirements of DCOI as follows, “The DCOI, as described in this memorandum, requires agencies to develop and report on data center strategies to consolidate inefficient infrastructure, optimize existing facilities, improve security posture, achieve cost savings, and transition to more efficient infrastructure, such as cloud services and inter-agency shared services.”
This initiative focuses on data center consolidation and optimization of existing data centers to reduce redundancy. These measures will help make data centers more eco-friendly which benefits the environment but by improving efficiency also provides significant cost savings. Further, DCOI recognizes and encourages the utilization of the cloud to improve efficiency and scale operations without expanding physical footprint. Undoubtedly, plans and goals will have to be established to meet the demands of DCOI so early adoption is the best approach. Schneider Electric explains the importance of complying with DCOI, “ One of the key requirements for existing data centers is to “achieve and maintain” a PUE score of 1.5 or less…new, proposed data centers must be designed and operated at 1.4 or less with 1.2 being “encouraged”. Another key requirement is deployment of data center infrastructure management tools (DCIM) in all Federal data centers since manual collection of PUE data will no longer be acceptable. If Agency CIOs fail to achieve these scores and implement DCIM by September 30th, 2018, “Agency CIOs shall evaluate options for consolidation or closure…”. In other words, comply or be assimilated. Fortunately for these CIOs, legacy data centers often have plenty of room to improve infrastructure efficiency by reducing power and cooling energy losses to bring PUE scores within these limits. In addition, DCOI targets are expected to result in the closure of approximately 52% of the overall Federal Data Center inventory1. So it’s important to try to make as many improvements as is feasible even if you’re meeting the required 1.5 (or 1.4 for new) …i.e., increase your odds of survival by being as good as you can be. Agencies should start with an efficiency assessment of the site in question. Find out where you’re at now and identify areas for improvement.” DCIM technology should be implemented to monitor energy usage and improve energy consumption and every effort should be made by federal data centers to comply with DCOI going forward.
Posted in Cloud Computing, computer room construction, Computer Room Design, Data Center Build, Data Center Construction, Data Center Infrastructure Management, DCIM, Power Management
Tagged computer room construction, computer room design, data center equipment, DCIM, DCOI
Every year we see certain trends arise in data center design and construction and, as 2016 winds down; we are able to take a look back at the year and anticipate what may be ahead in 2017. Data center design is constantly evolving as infrastructure changes and storage needs shift. Data center design is an expanding facet of the data center industry because every data center must be flexible and always constantly capable of adaptation and updating to stay current. Technavio points out just how important data center design is and will be looking forward, “This is why the global data center design market, which was valued at just $516 million in 2015, is expected to top $1.2 billion by 2020, growing at a cumulative average growth rate of 19.03%.” A data center that can scale on demand will be the data center that flourishes in 2017.
When it comes to data center design, one of the most common problems that data centers encounter is outgoing their existing space. This is combatted in a number of ways including increasing rack density, colocation, and usage of the cloud. Colocation will be a big trend moving forward for many businesses that have simply outgrown their enterprise data centers, or do not want to take on the task of managing IT infrastructure and protecting data security in a world where threats are constantly evolving. When businesses opt for colocation, it often opens the door to on-demand scalability and peace of mind that IT experts are managing data security. Additionally, we will see an increased use of the cloud in data centers to meet growing data demands. CloudTweaks offers a concise explanation of why the cloud will be transformative in data center design moving forward, “While organizations continue to consolidate facilities to save money, their need for effective data management and storage have increased exponentially. The volume of digital data is growing at an unprecedented rate, doubling every two years. Today’s IT execs are under phenomenal pressure to deliver value, while maintaining cost and efficiency. This is where the cloud can be most effective. Through economies of scale, cloud vendors are able to deliver the same, if not better, performance than in-house data centers at a lower cost. Furthermore, the cloud provides a centralized computing system that enables data and applications to be accessible from anywhere, anytime, yielding operational efficiencies.” Lastly, we will see a continued trend of making data centers more “green.” This means making efforts in design to more efficiently cool data centers and more efficiently supplying necessary power. These three trends are sure to be strong going into 2017 and will directly impact data center design.
Ask just about any client what one of the most important things they are looking for in a data center is and you will likely hear, “security” over and over again. Securing the traditional data center is a challenge unto itself but now many data centers are hybrid of traditional storage and cloud storage which complicates security immensely. Information Week describes the challenges faced in data center cloud security, a well as the strengths it will have moving forward, “Moving beyond traditional perimeter security into public, private, and hybrid cloud architectures stretches the capabilities of traditional security tools. It creates new security holes and blind spots that were not there previously. But cloud security is looking brighter by the day, and very soon cloud security tools will outmatch any type of non-cloud parameter security architecture. In many ways, cloud security is gaining in strength based on a seemingly inherent weakness. Cloud service providers are in a unique position to absorb vast amounts of data. Because large clouds are geographically dispersed in data centers around the globe, they can pull in all kinds of security intelligence as data flows in and out of the cloud. This intelligence can then be used to track security threats and stop them far more quickly.”
The problem is not a static one, it is a fast-paced, growing challenge. The more heavily the cloud is used, the more information it stores, the more security is needed but also the more potential holes in security there are. Security must scale at the same rate, or faster, than the growth of the data center. Because it is relatively new, and rapidly evolving, there are not clear-cut standards for cloud security in place. It is important for data centers to pay attention to what is happening in the industry and look to others, such as the U.S. government, for what is working in cloud security, which TechTarget elaborates on, “For example, cloud providers that handle confidential financial data should underscore their compliance with the Payment Card Industry Data Security Standard (PCI DSS) specification as proof of the integrity and security of their operations. PCI DSS does outline requirements related to cloud-specific aspects of security, stipulating that providers must segregate cardholder data and control access in addition to providing the means for logging, audit trails and forensic investigations. But the highly dynamic nature of most cloud-based applications — which often lack built-in auditing, encryption and key management controls — makes it expensive and impractical to apply the PCI standard to most cloud applications. Providers and enterprises seeking answers on cloud standards for security have found guidance from an unlikely source: the U.S. government. Though not usually perceived as a leading-edge technology adopter, the public sector has been engaged in aggressive security standards development efforts to support its Cloud First initiative, which requires federal agencies to select a cloud service for new deployments when a stable, secure and cost-effective offer is available. The Federal CIO Council laid out 150 cloud security controls for its Federal Risk Assessment Program (FedRAMP), which provides a baseline for common security requirements that agencies can use to verify that a prospective cloud provider supplies adequate cloud application security. Compliance will be validated by third-party assessment organizations. Using cloud-specific security requirements created by the National Institute of Standards and Technology (NIST), FedRAMP offers agencies a common set of cloud standards they can use to sanction a cloud provider. If the particular agency has additional security requirements, then the provider can build on its baseline controls to address these needs.” The a cloud is a cost-effective way for many data centers to scale to meet customers’ needs but security protocols must be in place to ensure that, as scaling occurs, data continues to be properly secured.
Data center power consumption is evolving all the time, becoming more efficient but, generally, growing. While many data centers are making green initiatives and finding ways to make their energy usage as efficient as possible, data demands are constantly growing, rack density is being increased and the need for effective cooling growing along with it. There are many approaches to data center cooling and even single data centers are implementing a variety of approaches to best cool their data center. Large data centers from companies like Yahoo or Apple are setting the trend in green data center cooling initiatives and small data centers are not only taking note but are also implementing those trends in their own data centers. Below are 5 exciting trends in data center cooling.
- Liquid Cooling
- Using liquid to cool, instead of air, is a great way to cool higher density racks and can be used in a variety of ways in the data center. TechTarget elaborates on the use of liquid cooling in data centers, “Now, new technologies can put 250 kW in a single rack, using liquid immersion cooling to play an important role for certain systems, such as high-performance computing, Cecci said. The pluses of liquid cooling include the ability to deploy it in specific areas — by row and rack — and it is very quiet and reliable, with few moving parts. Despite its benefits, liquid cooling is not in many data centers today, he said. “Most of these technologies — we will see them in the next two to three years,” Cecci said.”
- CRAC (computer room air conditioner) cooling systems has been used for a considerable amount of time in data centers. While they may be the old standard, CRAC has continued to evolve over time with new strategies to be an effective form of cooling in data centers, which TechTarget explains, “The easiest way to save money is to reduce the number of running CRAC units. If half the amount of cooling is required, turning off half the CRAC units will give a direct saving in energy costs — and in maintenance costs. Using variable-speed instead of fixed-speed CRAC units is another way to accomplish this, where the units run only at the speed required to maintain the desired temperature. The units run at their most effective levels only when they run at 100%, and some variable speed systems don’t run at a fully optimized rate when operating at partial load. Running standard, fixed-rate CRAC units in such a way as to build up “thermal inertia” can be cost- effective. Here, the data center is cooled considerably below the target temperature by running the units, and then they are turned off. The data center then is allowed to warm up until it reaches a defined point, and the CRAC units are turned back on.”
- Bypass Air
- Bypass air is any air that is conditioned that does not pass through IT equipment before returning to the cooling unit. In essence, it is a waste of cooled air which has led many data centers to make efforts to reduce the problem. Data Center Dynamics explains the problem and how data centers are fixing it to improve data center cooling, “The velocity of the cool air stream exceeds the ability of the server fans to draw in the cool air; as a result the cool air shoots beyond the face of the IT rack. Cool supply air can join the return air stream before passing through servers, weakening cooling efficiency. Eager to combat the inefficiencies above and keep pace with steadily climbing data center temperatures, businesses often adopt hot aisle/cold aisle rack orientation arrangements, in which only hot air exhausts and cool air intakes face each other in a given row of server racks. Such configurations generate convection currents that produce improved airflow. Although superior to chaos air distribution, hot aisle/cold aisle strategies have proven only marginally more capable of cooling today’s increasingly dense data centers, largely because both approaches ultimately share a common, fatal flaw: They allow air to move freely throughout the data center. This flaw eventually led to the introduction of containment cooling strategies. Designed to organize and control air streams, containment solutions enclose server racks in sealed structures that capture hot exhaust air, vent it to the CRAC units and then deliver chilled air directly to the server equipment’s air intakes.”
“Renewable energy.” “Clean energy.” These may sound like buzzwords – trendy little catchphrases meant to grab your attention and sound good but they are far more than buzzwords, they are the reality and the future of data centers. As we make pushes to become sustainable in all industries one of the biggest focuses will likely be towards sustainable, renewable energy in the data center. Large data centers use exponential amounts of energy, equivalent to the energy some small cities use, so it only goes to logic that there would be a push to make that energy usage as clean as possible. So, just how critical will it be that data centers focus on sustainability through renewable and clean energy going forward? Very. In fact, Data Center Knowledge notes that a study was completed and what consumers want now and going forward are data centers focused on sustainability, “A recent survey of consumers of retail colocation and wholesale data center services by Data Center Knowledge, found that 70 percent of these users consider sustainability issues when selecting data center providers. About one-third of the ones that do said it was very important that their data center providers power their facilities with renewable energy, and 15 percent said it was critical.Most respondents said their interest in data centers powered by renewable energy would increase over the next five years. More than 60 percent have an official sustainability policy, while 25 percent are considering developing one within the next 18 months.”
As data center space across the globe continues to rapidly grow, so will the amount of energy used. That energy use is often not only bad for the environment but quite costly. We have already seen large companies like Google and Apple focus on renewable energy and, as we often see, smaller data centers will likely follow in their footsteps. Small and large data centers are undertaking renovations and making changes towards renewable energy because even the tiniest improvements in efficiency and sustainability are saving big bucks. What do these renewable energy efforts look like? There are a vast array of options and approaches but Data Center Frontier elaborates on a few, “In broad terms, “clean” or “green” energy comes from renewable sources such as the sun (solar), wind, the movement of water in rivers and oceans (hydroelectricity), biofuels (fuel derived from organic matter), and geothermal activity. Today, there are big trends showing that tech giants are moving towards renewable energy sources in their green data centers. Digital Realty, along with certain major technology companies and other pioneers, are showing that clean energy can be used to power even the largest and most high-performance data centers. And as more organizations consider moving from traditional to cleaner sources of power, they are also showing that renewable energy can be cost-effective.” This is not a fleeting trend. Renewable energy is here to stay, it is the future of data centers, and all data centers should b emaking efforts to make small and big changes towards renewable energy for the future.
Every business can experience information and data silos on some level, particularly when various applications and systems must communicate with each other. But, these silos are particularly evident, costly, and problematic in data centers. Data Center Knowledge offers one example of the type of silo that can occur in a data center, “Electrical and mechanical systems in data centers are a perfect example of legacy IoT, he says, operating in silos, isolated from the IT systems they support. That isolation is the decades-old legacy, used to this day as the only method of securing these critical systems from intrusion.” Not only are there electrical and mechanical silos in the data center, but data/information silos as well. As certain tools and information are used in various ways by assorted applications, data silos emerge and become increasing problematic. So often, we see that when information silos occur, duplication of information and processes occur which takes up more space and thus uses more energy. Silos in data centers are truly a drain on resources.
The problem of data center silos is further exacerbated by the fact that, often, one company may have multiple data centers in locations all over the world. To avoid data center silos in both large and small data centers, there must be collaboration and open lines of communication. Fortunately, many data centers are opting for convergence rather than expansion, finding ways to use existing space in a more effective and efficient way. This alone, will help reduce information silos. Data Center Knowledge explains how convergence is being actively applied in data centers and elaborates on the advantages to be found when data centers opt for convergence, “As we saw with many Datalink enterprise customers who moved from the silo model, IT data centers first began to incorporate server virtualization technology to logically represent multiple servers on one or more consolidated, physical systems with smaller data center footprints. The data centers also began to incorporate their own dedicated network and shared storage to support them.Suddenly, one physical server could be used to serve up the needs of multiple applications, which it often did with glowing results. But, a ripple effect of virtual server growth often expanded storage and network needs significantly. Suddenly, cost savings in one area could be offset by growing expenses in another…
The benefits for IT can lead to:
- Less moving parts (and less individual vendor touch points) to manage or troubleshoot
- Greater resource utilization at a lower cost
- Faster application provisioning (one enterprise customer went from their prior three weeks to just 15 minutes to provision new applications)
- Faster IT response to business priority changes or changing market conditions
- Easier scaling and greater elasticity of the infrastructure
- Related integration and cross-training of previously siloed IT teams, themselves, in order to align IT further to the business
- A shorter pathway to on-demand services or private cloud environments to meet the IT needs of internal business units
A lower cost to support the growing data and application needs of the business (Another enterprise customer who runs its own SaS business found itself able to offer better quality services to current and new customers at a lower overall cost to itself.)”
Enterprise data centers may be a dying breed. Today we are seeing more and more data centers opt for colocation over enterprise data centers because of the high cost and level of expertise needed to run an enterprise data center. Additionally, cloud service providers are mitigating the need for enterprise data centers. Data Center Journal explains the basic appeal of colocation to enterprises, “Many businesses don’t have the time and money to invest in the equipment, technology, security and staff to run a full data center. For those businesses, colocation can help them optimize their department and free up resources, giving employees the time and bandwidth to focus on more strategic business tasks. Colocation facilities provide the space, cooling, power and security for your server, storage and networking equipment, while giving IT manager’s access to high bandwidth, low latency and always-on connections… Large colocation facilities also offer significant benefits of scale. By utilizing large power and mechanical systems, the facility can provide high uptimes and speed as well as the ability to efficiently appropriate additional resources and quickly grow alongside your company.”
Colocation is certainly not free, it comes at a cost (particularly at the beginning) and provides its own set of security risks. But, while enterprise data centers will never completely disappear but they are certainly fading, and rapidly. Data Center Knowledge discusses the trend of moving away from enterprise data centers towards colocation and cloud services, “As Liz Cruz, associate director with the market research firm IHS and the panel’s moderator, pointed out, hardware and infrastructure equipment sales into data centers are declining, while revenue colocation providers are raking in is growing in double digits, which means more and more companies choose outsourcing over their own data centers. Still, when she asked people in the audience to raise their hands if their companies had at least two-thirds of their IT capacity in colocation data centers, only a handful did. It’s cloud providers who are driving a lot of the revenue growth for colo companies – a lot more than enterprises, although enterprise data center spending is slowly waning. “Cloud providers are now the largest tenant of multitenant data center facilities,” Cruz said… For colocation providers, these hard-nosed enterprise users are not only a big growth opportunity; it’s a matter of longevity. The race to capture the hearts and minds of the enterprise is on, but they’re not only racing each other. They’re also racing the Amazon, Microsoft, and a few others. Most colo providers have embraced public cloud as reality and have been using their ability to provide direct network access to cloud services from their facilities as a way to attract enterprises, pitching customers on the hybrid cloud, where a physical footprint the customer has full control of is supplemented with public cloud services, all under one roof in a colocation facility.” If the trend continues, as we suspect it will, colocation data centers will continue to grow and work towards integration with cloud services to draw more and more businesses away from enterprise data centers.
The tech industry loves to uses catchy phrases to describe various processes, innovations and aspects in data centers. Every now and then, we think it is important to narrow in on those phrases and explore what they mean and how they impact data center operations. One of those phrases is “data in motion vs. data at rest.” Data in motion vs. data at rest are somewhat self-explanatory but their nuances and impact on data centers are not. Data in motion is data that is actively being used by data centers, it is data in transit. Data at rest is data that is not being actively used but is stored in a data center. These two different types of data present unique security challenges. For example, data in motion may be in transit on the internet and that presents different security challenges than data at rest that, while not actively in use, may contain sensitive customer information.
When it comes to securing data of any kind there are a variety of ways to prevent security breaches and cyber-attacks. Encryption is certainly a must and data at rest must be secured on a number of levels because it may be stored in multiple places, including databases, storage networks, file servers, or virtually in the cloud. Data in motion is exposed to cyber-attacks at a number of points while in transit. DataMotion explains what risks may be encountered and describes some best practices when it comes to securing data, “Data is at rest when it is stored on a hard drive. In this relatively secure state, information is primarily protected by conventional perimeter-based defenses such as firewalls and anti-virus programs. However, these barriers are not impenetrable. Organizations need additional layers of defense to protect sensitive data from intruders in the event that the network is compromised. Encrypting hard drives is one of the best ways to ensure the security of data at rest. Other steps can also help, such as storing individual data elements in separate locations to decrease the likelihood of attackers gaining enough information to commit fraud or other crimes…Data is at its most vulnerable when it is in motion, and protecting information in this state requires specialized capabilities. Our expectation of immediacy dictates that a growing volume of sensitive data be transmitted digitally— forcing many organizations to replace couriers, faxes, and conventional mail service with faster options such as email. Looking ahead, it will also become increasingly important for the encryption service your organization uses to cover mobile email applications. The Radicati Group1 predicts that 80% of email users will access their accounts via mobile devices by 2018, but more than 35% of organizations currently using email encryption say their users currently lack the ability to send secure messages from their mobile email client.” Securing data will be a challenge that will never leave, data centers must look forward, anticipate potential future threats, and use multi-level encryption to ensure that data in motion and data at rest remain protected.
From time to time we see new “catchphrases” or terminology pop up in the tech world and suddenly they are being used everywhere. One of these phrases is “the internet of things.” As we see our world become increasing automated, digitized and more, we see that many aspects of our day to day life are now controlled by or taking place on the internet. Forbes explains exactly what “the internet of things” means, “Simply put, this is the concept of basically connecting any device with an on and off switch to the Internet (and/or to each other). This includes everything from cellphones, coffee makers, washing machines, headphones, lamps, wearable devices and almost anything else you can think of. This also applies to components of machines, for example a jet engine of an airplane or the drill of an oil rig. As I mentioned, if it has an on and off switch then chances are it can be a part of the IoT. The analyst firm Gartner says that by 2020 there will be over 26 billion connected devices… That’s a lot of connections (some even estimate this number to be much higher, over 100 billion). The IoT is a giant network of connected “things” (which also includes people). The relationship will be between people-people, people-things, and things-things.”
The reality is, whether people like it or not, the internet of things is taking over and our world is being powered by the internet. This impacts our day to day life in different ways but one thing data centers know is that it means more data. Everything that is becoming part of the internet of things involves data. That coffee machine that is connected to the internet is going to require data communication and storage on some small level. And, if a jet engine is connected, it is probably using a lot of data. As more things become part of the collection of the internet of things, data center demands exponentially increase. Data centers must begin to prepare now because as tie marches on, the internet of things will only increase. Data Center Dynamics points out just how much this will impact data centers going forward, “The internet of things will force enterprise data center operators to completely rethink the way they manage capacity across all layers of the IT stack, according to a recent report by the market research firm Gartner… Where this becomes problematic for data centers is management of security, servers, storage and network, Joe Skorupa, VP and distinguished analyst at Gartner, said. “Data center managers will need to deploy more forward-looking capacity management in these areas to be able to proactively meet the business priorities associated with IoT,” he said in a statement. For data center networks, the internet of things will basically mean a lot more incoming traffic. WAN links in data centers today are designed for “moderate” bandwidth requirements of human interaction with applications.Data from multitudes of sensors will require a lot more bandwidth than current capacity… Of course a lot more data will mean a lot more storage will have to be provisioned in data centers. In addition to pure capacity, companies will have to focus on being able to get and use data generated by the internet of things cost effectively.Because of the volume of data and the amount of network connections that carry it, there will be more need for distributed data center management and appropriate system management platforms.” While end-users may take for granted the convenience of the internet of things, data centers do not have the luxury of taking it for granted. They must be vigilant in preparation and expansion to ensure they can accommodate the dramatically growing data needs that the internet of things presents.
If there is one thing that a data center is concerned with, aside from maximizing uptime, it is security. In today’s world we constantly hear news stories about security breaches exposing businesses and individuals to danger such as identity theft, information loss, other theft, and more. Security breaches are not just an embarrassing frustration; they are a costly one as well. Large businesses can obviously suffer significant losses but the losses experienced by small and medium-sized businesses are significant as well. Security Intelligence describes the growing risk of security breaches, “Every corner of the organization — from human resources to operations to marketing — is generating, acquiring, processing, storing and sharing more data every day. Cybersecurity threats have conditioned organizations to defend the full depth of this sensitive information and infrastructure from a global threat landscape…IBM and Ponemon Institute are pleased to release the “2015 Cost of Data Breach Study: Global Analysis.” According to our research, the average total cost of a data breach for the participating companies increased 23 percent over the past two years to $3.79 million.” This growing problem and increasing cost is a clear signal that data centers and business must all pay careful attention to security measures to ensure that data is properly protected.
While upping your security protection will certainly involve an up-front investment, if you are protecting critical information such as health, financial, social security, or other top secret records, the cost of a breach will be far more than the cost of protection. For example, cloud security may be ideal for less critical information (i.e. social media) but higher security protection is better for more sensitive information. If you have a smaller business and data center, you may think the risk of a security breach is smaller, statistics are showing that, in general, security breaches are a growing reality for many. Data Center Knowledge points out the frequency of security breaches,” Roughly half of businesses in the U.S. (49 percent) and globally (52 percent) assume that their IT security will be breached sooner or later. This is a recognition of reality, as 77 percent of U.S. businesses and 82 percent globally have experienced between 1 and 5 separate data security incidents in the last year.” Data Center Knowledge also notes that smaller and medium businesses that experience a security breach typically incur a loss, on average, of $86,500. And the cost of liability for data breaches is growing, emphasizing the importance of protecting their customer’s and user’s private information, which Data Center Knowledge points out, “There’s legislation brewing that would make organizations far more accountable for breaches of personal information and require them to pay actual damages to individuals, something he thinks will reverse the trend toward cloud and colocation back to in-house.”