The turn of seasons brings about many changes. We adjust the thermostats in our homes and, similarly we must also control the environment in a data center. But, it is not just about accommodating for the varying temperature and adjusting the thermostat. In a data center, humidity must also be accommodated for to keep equipment functioning properly. While temperature and cooling is often the focus, humidity control cannot be overlooked. There is a range of acceptable humidity at which most data centers should maintain but if they stray outside the range, problems can ensue.
Data centers must constantly monitor the humidity level in its room. When a data center is too humid condensation will inevitably show up. If condensation happens on the many electrical components in a data center they could possibly short out. Shorting out equipment = big problem for data centers. Conversely, when humidity is too low ESD or electrostatic discharge, can occur. And yes, it is as frightening as it sounds. ESD emits a static electricity shock and when that happens in a data center it could shock electrical equipment so much that it shuts down completely. The shock may even be so strong that it damages electrical equipment. All of these scenarios lead to downtime in a data center and as any data center manager knows, downtime costs money and leads to a lot of frustration, even if downtime only lasts a moment. For this reason it is critical that data centers implement a humidification system. Once humidification systems are implemented it is critical that they are maintained on a regular basis to ensure that they are working year round. Just like any other equipment in the data center, humidification systems can gradually wear down over time or encounter seasonal glitches, right as they are most needed. An annual, or better yet, bi-annual check of the humidification system will help ensure that it is working when it is needed most so that mission critical data center equipment can be protected. Humidity tends to decrease during the colder months and increase during the warmer months so appropriate adjustments for those seasonal changes can be made in advance. 40% humidity is around the recommended level of appropriate humidity for a data center but the acceptable range is much larger – between 20-80%. Most experts say that maintaining closer to 40% is truly ideal, though, and that by doing so equipment will be better protected and the heating and cooling system will not need to work as hard to maintain appropriate temperatures.
It is the classic dilemma for data centers – move or renovate? When technology moves a mile per minute and everything from applications to infrastructure is constantly evolving a data center must evolve along with it or it will find itself outdated, inefficient or in need of a move. But, how do you know when it is time to renovate? It is important for data center managers to keep a good grasp on energy efficiency and infrastructure with a well functioning DCIM plan. This way, data center managers will be ahead of the game and be able to anticipate major problems before they even arise. Moving a data center is a major undertaking so watch out for the signs below and consider renovating your data center rather than moving.
Three Signs It Is Time to Renovate Your Data Center
- Technology Lag
- Data centers are nothing if not technologically driven so if the technology in a data center is behind the times it is bound to create a variety of problems. From issues with application compatibility to infrastructure compatibility to all-around inefficiency a technology lag is a sure sign that it is time to renovate a data center.
- No Space
- One of the biggest problems that any data center can run into is a lack of space. Unfortunately, this happens all too often. You add a rack here or there, some more infrastructure and all the sudden you realize that you have run out of space or will very soon. You are faced with the classic data center dilemma – renovate or move? Moving is incredibly costly and tricky while also trying to maintain uptime. So, rather than move, many data centers should consider renovating so that they can continue to use their existing location while increasing their usable space.
- Energy Bills Are Skyrocketing and So Is the Temperature
- If you feel like you are walking on the surface of the sun while you walk around in your data center that is probably a big red flag. And, if this is happening you are probably noticing that energy consumption and thus, energy bills are increasing. When you have not implemented proper heating and cooling techniques for your data center to ensure the most energy efficiency possible it is time to stop what you are doing and do so. Data centers have a lot of server racks and server racks generate a lot of heat. Without proper cooling solutions and energy management a data center will not run properly so the sooner you renovate to improve heating and cooling the better.
Data center energy efficiency is at the forefront of hot topics for data centers. And, for good reason. Data centers use a truly astonishing amount of energy each year. The Natural Resources Defense Council (NRDC) noted just how much energy data centers use, “In 2013, U.S. data centers consumed an estimated 91 billion kilowatt-hours of electricity, equivalent to the annual output of 34 large (500-megawatt) coal-fired power plants (and, the NRDC notes, the equivalent of enough electricity to power all the households in New York City twice over). Data center electricity consumption is projected to increase to roughly 140 billion kilowatt-hours annually by 2020, the equivalent annual output of 50 power plants, costing American businesses $13 billion annually in electricity bills and emitting nearly 100 million metric tons of carbon pollution per year.” Improving energy efficiency in a data center is incredibly important but can seem like a daunting task for many data center managers who often are not even certain where to begin, if they have the budget to make changes, or if the changes they make will truly make an impact.
One of the first, and most practical ways to begin a shift towards improving energy efficiency is to take a real look at energy usage. What is using the most energy, should it be using that much energy and can anything be eliminated? Often, energy is being wasted on ghost infrastructure or outdated energy-draining equipment. But, as many data center managers know, it can be difficult to keep track of all of the infrastructure in a data center or truly know what is using the most energy of inefficiently using energy. That is why a good DCIM plan is important so that data center managers can work with the most current information rather than outdated information and make well-informed decisions going forward. Once you have sufficiently audited data center energy usage and are able to make well-informed decisions for improvement you can move onto the next step. Next, make immediate changes to improve energy efficiency while also looking at long term improvements. Long term improvements are incredibly important and it is wise to look at how to remain sustainable in the future but while making decisions for the future you can make some immediate changes such as hot aisle/cold aisle arrangements or other containment options that may help improve energy efficiency. While other improvements are more difficult or costly to implement, containment arrangements can be made and executed relatively quickly and will make a big impact. While making immediate changes it is important to get budgetary approval for bigger changes and, once approved, begin moving forward with changes that will help your data center remain efficient in the future. This most likely means upgrading equipment to the most current, energy efficient options. All equipment has a lifespan and once they are getting a big old they will likely become energy inefficient. If you have budgetary approval to make improvements with heating and cooling options as well it is a great choice to improve your cooling ability within a data center because cooling is typically one of the biggest expenses in a data center. Lastly, explore green options like making a switch to more fully utilizing cloud storage or implementing cooling with outside air sources so that you can be as energy efficient as possible now and in the future.
Data center energy consumption is a major topic of conversation. Data centers are one of the largest consumers of energy in the country and in the world. Data center energy consumption can exceed that of power plants and, therefore, must be managed carefully and optimized whenever possible. But, the CFO may be difficult to get on board when it comes to energy efficiency measures. The financial investment necessary to improve energy efficiency in many data centers can often seem astronomical. While it may be a significant budgetary expense, it will be worth it. The task of the data center manager and IT head is to convince the CFO that it IS necessary, and why. Energy consumption always seems to outpace efficient energy management. It is a never-ending headache and frustration for data center managers because optimization is important for a data center to be more eco-friendly and save money. While data centers of the past may have been energy guzzlers, today’s modern data center must be optimized for ideal power usage and there needs to be room in the budget to make it happen.
To properly manage data center energy consumption, a thorough, evolving, and future-proof DCIM must be in place. The infrastructure of a data center is, after all, a big portion of data center energy use. The PUE must be analyzed and noted because it is an easy way to show a CFO that your data center is not efficiently using power. Increasing rack density and growing infrastructure tend to also increase cooling needs and suddenly, energy is not being managed properly and is being wasted. Rack density gets increased because of the growing needs of the data center and the amount of data that needs to be stored. If your equipment is out of date it is probably consuming far more energy than you realize. While updating equipment is a big financial endeavor on the front end, it will pay for itself significantly on the back end because your data center will run more efficiently and use less energy. The next important focus for data center managers when optimizing power usage is to look at the data center’s physical setup. Is it optimized to make cooling as efficient as possible? If your data center is housed in an older building that has not been properly renovated it probably is not optimized. Consider efficient cooling methods that can reduce energy used trying to cool an inefficient environment. Hot aisle/cold aisle technique is one way to improve cooling efficiency, other containment options such as a ceiling-ducted air containment, cold rooms and more are other options that data centers are taking advantage of to keep their data centers cool while still maintaining energy efficiency. The CFO may not have much to do with the day to day operations of a data center but a data center renovation or move cannot take place without the budgetary approval from a CFO and therefore DCIM analysis and management must be done often so that a detailed picture of data center energy use can be presented and efficient change can begin.
Technology continues to deeply integrate itself into our everyday lives. Wearable technology is quickly becoming commonplace, something you see on everyone. Business Insider reports on just how rapidly wearable technology is growing, “In just a few years, there could be more people using wearable tech devices than there are in the US and Canada. In a note to clients on Monday — alongside initiation of Fitbit coverage — Piper Jaffray’s Erinn Murphy and Christof Fischer stated that “wearable technology will be the next generation of devices to transform how individuals consume and use information.” Murphy and Fischer estimate the wearable tech category will grow from 21 million units in 2014 to 150 million units in 2019, a 48% compound annual growth rate (CAGR). This growth will largely be fueled by wrist wearables like smart watches and fitness bands.” Typically, when you think of the cloud you probably think of data centers managing vast amounts of information, storing and analyzing data to better improve business practices and providing an important efficiency improvement for data storage. While all of these things are accurate, those kinds of things do not really directly touch the consumer in a tangible way. But, as technology continues to evolve and more and more wearable technology enters the marketplace, the cloud is quickly integrating itself into the consumer’s daily life.
When one looks at something like Google Glass, the Apple Watch, or the Fitbit, they probably do not immediately think of the cloud. But, the cloud is an integral part of wearable technology. Why? because people are not simply interested in cool technology, they want that technology to actually improve their lives, make things more simple and provide a service. This cannot be accomplished without the cloud because the cloud collects the data from these wearable technologies, stores the data, analyzes the data and then uses the data for the benefit of the wearer. This is what makes wearable technology so exciting and desirable to consumers. Wearable technology, in essence, performs a service because it collects the data and then stores, analyzes or uses it in a beneficial and helpful way. These technologies are not simply for entertainment and personal use but also for the workplace as well. Because wearable technology can also improve careers and facilitate better work it makes them even more valuable. The better the cloud service, the more accurate the data collection and analysis is, the better the overall product will be, and thus, the more successful and marketable the product will be. Therefore, it is critical that quality cloud service work closely with wearable technology as the world becomes more and more dependent on the services provided by and features of wearable technology.
Few things in recent history point to the need for a comprehensive disaster recovery plan for data centers than Hurricane Sandy. When disaster struck many data centers were unprepared and ill-equipped which led to significant downtime that ultimately cost millions of dollars. Data centers can only operate for so long under the false sense of security and protection before a disaster strikes and the sudden panic sets in. A detailed disaster recovery plan must be in place, complete with multiple contingencies, before a disaster ever strikes so that, should it happen, immediate action can be taken.
To begin formulating a disaster recovery plan you must first identify all of your critical systems. Once you have identified them, you can properly determine how to best protect them in the event of a disaster. To properly prepare a detailed inventory of infrastructure, along with a comprehensive understanding of it must be routinely kept. When DCIM is lagging and knowledge of infrastructure lacking or out of date a disaster will become a major problem for data centers. Along with a detailed knowledge, a thorough backup must be in place as well. Redundancy in a data center not only protects against lost data on a day to day basis but in the event of a disaster as well. Additionally, by taking advantage of the cloud data centers can virtually protect information which is safe from disaster and a useful tool for data recovery. Ready, a national public service campaign that is “designed to educate and empower Americans to prepare for and respond to emergencies including natural and man-made disasters,” describes what the critical elements of a disaster recovery plan are for any data center, ” Information technology systems require hardware, software, data and connectivity. Without one component of the “system,” the system may not run. Therefore, recovery strategies should be developed to anticipate the loss of one or more of the following system components:
- Computer room environment (secure computer room with climate control, conditioned and backup power supply, etc.)
- Hardware (networks, servers, desktop and laptop computers, wireless devices and peripherals)
- Connectivity to a service provider (fiber, cable, wireless, etc.)
- Software applications (electronic data interchange, electronic mail, enterprise resource management, office productivity, etc.)
- Data and restoration
Once you have identified critical system components and how to best protect them, a comprehensive DR (disaster recovery) Plan should be formally written up and kept safe. All pertinent personnel should also be trained and prepared for how to act should a disaster occur. By doing so, you will be able to best protect your data center and clients if a disaster occurs and maximize uptime.
It is critical that any data center have a strong DCIM strategy in place. Even in relatively small data centers the amount of infrastructure that must be tracked, managed, maintained and more is enough to overwhelm any data center manager. There are a variety of ways to track and maintain infrastructure but traditionally it has been done manually. But, is this method truly the most effective way to collect data in a data center? And, if not, what is the solution?
One way that many data centers have traditionally collected data is to give each piece of infrastructure a barcode. The barcode can then be scanned and inventoried to keep track of what is where. But, the problem with this method is that it does not provide much data and the data it does provide is often outdated. Additionally, it does not provide much insight as to how each piece of equipment within an infrastructure is using energy. Without that information, how can a data center manager accurately determine where energy is being used appropriately, where it is being underused and where it is being overused? If an analysis is only completed every 6 months or once per year, that is a lot of time that has been wasted when improvements could have been made, efficiency improved and money saved. By implementing a more intelligent DCIM management tool it can provide up to data information about equipment and energy consumption so that it can offer real, actionable data that data center managers can use. Alerts can be arranged based on predetermined thresholds, graphs and charts generated from gathered data and more accurate action can take place. It is also a far more efficient means of collecting data in a data center rather than manual data collection. This will save time and many headaches for data center managers, making an improvement any way you look at it. One of the last and most important reasons to implement an automated data collection method instead of manually collecting data is that it can alert you to potential capacity problems far sooner than manual methods. Rather than capacity problems sneaking up on data center managers, leading to major problems and the possible expense and frustration of having to relocate, automated DCIM will help data center managers see where capacity stands within a data center far before outgrowth ever becomes a problem so that any necessary condensing, rearranging or cloud storing can take place to save room. Manual data collection may have been the way thing were done in the past but it is not the way of the future and data centers that make the switch now to automated DCIM will enjoy the variety of benefits it has to offer.
In today’s technology world everything is getting smaller and smaller. And, with the cloud, some things are simply disappearing! Many data centers struggle to make their capacity work without exceeding it but, as expansion occurs, infrastructure changes and more racks are needed, data centers often find themselves outgrowing their space. This leaves them in the frustrating position that they must then relocate which can be tricky and often leads to downtime. Racks-on-chips have become a hot topic of conversation because it can help slow down overgrowth in a data center. Racks-on-chips are essentially a condensed version of a server rack while maintaining processing power and capacity but packaged in a much smaller package – a chip.
The critical component of racks-on-chips that makes them possible is that the capacity of chips must be increased so that they can store as much as server racks can. Racks-on-chips can be achieved but data centers must be retrofitted to make them possible because racks-on-chips have to be networked together with optical circuit switching and electronic packet switching. New data center builds that are forward thinking will plan ahead and have these system in place so that racks-on-chips will be possible. By switching to racks-on-chips the amount of server racks used in a data center can be dramatically reduced and once of the biggest benefits of this aside from freeing up physical space in a data center is that is also significantly reduces the amount of energy needed for heating and cooling. This means that a data center using racks-on-chips will be far more energy efficient, which is not only more green but will save a significant amount of money. While racks-on-chips are a potential ideal solution for data centers space and energy issues, Motherboard explains that it is still in the early stages and much still needs to happen for the benefits to be truly realized, “But the meat and potatoes of Yeshaiahu Fainman and George Porter’s server-rack-on-a-chip vision is really about taking the existing framework for a server rack and recreating it at the nano-level. They say that miniaturizing all server components so that several servers can fit onto a computer chip would increase processing speed. Making circuit systems to support all these mini-components using advanced lithography is already feasible, but scientists have yet to realize nano-transceivers and circuit-switchers—the key components that transmit data. And while silicon chips are increasing being used to transmit data-carrying light waves in fiber optic networks, efficiently generating light on a silicon chip is still early in its development. The researchers offer some solutions, like including light generating nanolasers in the chip design.”
DCIM, or data center infrastructure management, is incredibly important for any data center. If infrastructure is not managed properly in a data center many problems can arise. Proper management is challenging for even the most experienced data center managers because technology is constantly evolving, power density is constantly added to racks and cooling needs shift as well. One of the best ways to improve data center infrastructure management is to utilize software to collect information rather than manually collecting information. DCIM software is a central location to aggregate and analyze data that has been collected so that performance and uptime can continue to be improved.
To properly manage capacity, data about the following must be collected: power usage, rack space, cooling and more. As this data is collected data center managers will be able to see how much rack space is being used so that they can determine whether or not they want to add more density to a rack. With this data, managers will know if rack density maximum capacity is being used properly or too close to being exceeded. If it is being underutilized too much energy may be used to cool it and it is likely better to increase density or condense server racks to maximize space and efficiency. This is important because proper management of rack density allows for improved energy efficiency and a reduction of downtime. But, without a sophisticated DCIM software, a data center manager may not even realize if a rack density is being under or over-utilized. Often, what data center managers experience is that, over the life of a data center, more systems are added, equipment is added or adjusted and suddenly there is a general lack of knowledge about power density. DCIM software helps prevent a lack of knowledge from occurring by ensuring that data center managers, and support staff, all have the most current information about what is happening in a data center. DCIM software will not only analyze rack utilization but power management as well. As data is collected data center managers can see how and where power is being used. This will help pinpoint areas in need of improvement and allow managers to make changes as quickly as possible. DCIM software can even be set up with automatic notifications so that the most current information is being provided at all times. With such current information it is easy to micromanage or become overly concerned with finite details but it is better to instead look at trends over time. By analyzing charts and graphs to see trends you will be able to see if something was a fluke or if it is a consistent pattern. Once consistent patterns are known it is easier and more effective to make permanent changes and improvements to a data center so that efficiency and data management can be optimized.
Many data centers operate with high density server racks for a variety of reasons. Perhaps they want to maximize their space by increasing the density of their racks so that they do not outgrow their data center walls. Perhaps it is a strategy to concentrate power, and therefore heat, in an attempt to more specifically cool areas as an attempt at improved cooling efficiency. No matter the reason , many data centers have extremely high PDUs, sometimes without even realizing it, and without realizing it, it can be incredibly challenging to appropriately cool a data center. This dilemma is often the source of outages and downtime, two things no data center wants to experience. No doubt about it, the pressure is on data center managers to run a high density data center that are also highly efficient and able to sustain their power needs. No small or simple task by any means.
So, what is a data center manager to do? How can you achieve high density, high efficiency and high efficacy all at once? Is it even possible? Yes, but data center managers must make it their priority to have a full working knowledge of their data center, power density capacities, best practices when it comes to cooling and more and then devise a strategy with contingencies so that if the unexpected arises there is a plan in place.
First and foremost, a data center manager must have knowledge of what a data center’s rack power density maximum capacities because without this knowledge first a data center will always be playing catch up. Catch up leads to outages and downtime so it is simply unacceptable. Once a data center manager has knowledge of power density maximum capacity they can work to ensure it is not exceeded. This alone will prevent a significant amount of problems. Additionally, by understanding maximum capacity a data center manager can make appropriate cooling arrangements. They will also be able to accurately determine what backup power and redundancies are needed to mitigate downtime. With a full working knowledge of data center power densities and cooling needs, alternative cooling methods can be explored to determine whether or not they would be more efficient or effective. If power density maximums are being pushed or exceeded additional server racks or space may be needed because pushing or exceeding the limits may temporarily save space but it will create major problems going forward. Outages, downtime, damage and more could be experienced if power density maximums are exceeded so the temporary solution was never really a solution at all. Once a strategy and plan is in place, the entire data center team and all other appropriate personnel should be well informed and trained because without communication problems will continue to arise. High density racks are a good strategy for data centers who want to maximize space and efficiency but for it to be effective a full working knowledge of data center abilities and capacities must be known and appropriate cooling methods must be executed.
Posted in Back-up Power Industry, computer room construction, Computer Room Design, data center cooling, Data Center Design, data center equipment, Data Center Infrastructure Management, DCIM, Power Distribution Unit
Tagged data center cooling, data center design, data center equipment