Data center capacity management is a hot topic among data center managers and staff. Effectively managing data center capacity is a challenging undertaking and one that is constantly evolving which keeps everyone on their toes. Without a capacity management plan that involves the entire data center team, as well as regular auditing of infrastructure, usage and capacity, a data center could quickly run into problems.
One of the best ways to help keep data center capacity management under control is to be well informed. How does a data center manager and team go about doing this? Regular data center capacity audits. As soon as you begin audits you will establish a baseline and as you evaluate your audits each month you will begin to see trends, ways to improve and be able to anticipate whether or not you may exceed capacity, and when. Your baseline will include information about how your storage, server, and network infrastructure are used. Within your data center infrastructure management (DCIM) everything is monitored and data is collected to allow for improved operational support and more efficient capacity planning. Capacity management is becoming more and more important as an emphasis is put on lowering data center expenses and improving efficiency. The logical step for many data center managers is to add more to their existing load within a data center rather than expanding or changing locations. But, the more you add the more your infrastructure must be capable of handling, and handling efficiently. When you can assess trends and more accurately predict future capacity needs it allows for better budgeting accuracy. This will help data center managers better manage expenses, investments, changes in infrastructure and more. This approach improves operational transparency and allows data center managers to anticipate future needs rather than react to them as they occur.
Of course, another part of capacity management is to establish plans. It is great to use data collected from analyzing usage, manage capacity and plan for future capacity needs but, beyond that, data center managers must formulate various plans for different potential scenarios and inform all data center employees of their roles in said plans. You cannot plan for just one scenario because, unfortunately, in an ever-evolving world, scenarios can change very quickly. It is simply best practice to plan for multiple contingencies and involve the entire team so that when it is time to execute plans it can be done efficiently and effectively. This form of anticipation, planning and execution is exactly the kind of capacity management that prevents downtime from occurring and allows a data center to run as it should at all times.
A lot of data centers are doing it but is it right for you? Relocation – how do you really know when it is time? For data centers, there may be some tell-tale signs that it is time to relocate. And, if your business is being relocated a data center or computer room may simply have no choice but to relocate. Relocating a data center is no small task and requires a good deal of planning and preparation so that it can be executed smoothly, downtime can be avoided and important data protected.
A data center manager needs to be on the lookout for signs that it may be time to relocate so that potential problems can be avoided. One of the first signs it may be time to relocate are a wide array of power problems. Whether your data center is no longer energy efficient (especially in spite of energy efficient improvements) or your power source is unstable or unreliable, it is probably time to relocate data centers. Energy usage, energy efficiency and the cost to power a data center are major topics of conversation for any data center and the bottom line is that if you cannot achieve energy efficiency within the existing data center, it will dramatically impact the bottom line and cost more money than it should. A data center designed for energy efficiency will not only be more eco-friendly but will save money which is important to any business. Additionally, if your power supply is unreliable that is just unacceptable in the data center business. If the power supply is not working and your backup power fails for some reason your data center could experience small amounts of downtime or, worst case scenario, significant downtime. Even small amounts of downtime come at a significant expense and prolonged downtime could mean lost revenue, frustrated customers and ultimately could cause a business to have to shut their doors. If your data center has any form of consistent power problems, it is time to relocate. The next problem that could indicate it is time to relocate is that there is no room to grow. As data centers grow in capacity and add more server racks and other assorted equipment spare room could very quickly run out. Once you have outgrown your space there is not much that can be done which means that it is probably time to relocate. When choosing a location, do not simply accommodate growth that has recently occurred but anticipate additional growth that could occur in the future so that you will not have to move again right away. An additional reason to consider relocating, especially if you have any of the aforementioned concerns, is if your location is not right. Wrong location could translate to a few different things – it may be inconvenient or too remote, it may be in an area where real estate is too pricey or it may not be in an area near the target market that you serve. If any of these are true they are probably frustrating and expensive issues and the best way to solve a location problem is to relocate. Determine if the benefits and potential savings outweigh the costs and difficulty involved with relocating a data center but a move may be just the right thing to mitigate problems for your data center.
Posted in computer room construction, Computer Room Design, Construction Industry, Data Center Build, Data Center Construction, Data Center Design, Datacenter Design
Tagged compute room construction, computer room build, computer room construction, Data Center, Data Center Construction, data center design
When you think of a healthcare facility and the IT challenges it may face, you probably first thing of large hospitals. And, while they certainly face a number of IT challenges, many other healthcare facilities, such as clinics, outpatient practices, dental offices and more face IT challenges as well. Today, medical records, patient files, physician notes, prescription orders and more are all kept on secure servers and transmitted securely via electronic file sharing methods. Paper files are quickly becoming a thing of the past and this means that all healthcare facilities, large or small, have to find efficient and effective means of managing their IT. Many small healthcare facilities are short on space as it is so a dedicated server room is likely not practical or even possible. Rather than having a dedicated server room, many healthcare facilities could simply opt for a server-room-in-a-box instead.
A server-room-in-a-box is small but still effectively houses IT equipment in one location that is ideally designed for security and energy-efficiency. A server-room-in-a-box can be housed in furniture that coordinates with the rest of the decor in a healthcare facility so it will blend right in. The housing can be on caster wheels or can be completely locked down depending on each healthcare facilities specific needs. When you hear about a server-room-in-a-box you may worry about overheating and energy efficiency but a server-room-in-a-box is self-cooled with fan options and ventilation so that you do not need any special equipment to keep it running properly. Also, if you have been in a server room you know that it can be noisy at times. With a server-room-in-a-box it is actually quiet because it will muffle noise from the equipment. Ideally you should aim to muffle as much noise as possible (aim for approximately 90%) so that you can create an ideal healthcare facility setting and work without interruption. The server-room-in-a-box you choose should have an integrated PDU to make power distribution easy. If you are also wondering if the capacity of a server-room-in-a-box will be enough, they actually come in a variety of capacities ranging from 12U to 38U. And, if necessary you could always add an additional server-room-in-a-box should capacity needs increase. Healthcare facilities face a number of different challenges than a typical office setting might face. It is not just employees visiting the facilities each day but patients and outside personnel as well. With that in mind, as well as reduced space needs, a server-room-in-a-box is an ideal approach to managing IT needs in a healthcare facility.
Today, many data centers are finding that, rather than trying to make an existing location work, it is better to consolidate data centers or move to a new location. Moving to a new location provides the opportunity to have more room, better energy efficiency and an optimized environment for any center needs. Any time a data center faces a big move, it will undoubtedly cause some stress and concern. After all, a move so large poses a lot of potential risks when it comes to data loss and downtime. Whether a data center runs critical components of a retail location, retail websites, healthcare systems, warehouse systems or any important operations that would impact staff and personnel if down for very long, downtime must be minimized to protect a business form significant lost revenue, frustration and major problems.
It can be difficult to properly assess the heating & cooling needs as well as the power and capacity needs of an existing data center for a data center manager. Because of this, when a data center is being moved, anticipating possible problems or needs during a move to be properly prepared can prove incredibly challenging. One problem many data centers encounter during a big move is miscommunication or lack of communication altogether. Operational information silos are not both unproductive and detrimental during a data center move. If communication breaks down important messages may not be relayed, things may not be done as they should, and downtime while everything is being sorted out is likely to occur. To avoid this, it is important to have all important people involved in a data center move involved in every meeting. Even if it seems that some may not be necessary for all meetings, have them be present so that everyone is on the same page. The next problem many data centers experience during a move is a lack of proper planning. When planning for a data center move it is important to anticipate all needs as well as possible unforeseen needs. Have contingency plans in place for the unexpected so that, should something unforeseen arise, you can quickly and efficiently mitigate the problem without loss of critical information or downtime. Next, it is wise to test critical data center equipment and infrastructure to ensure that everything is in proper working order and current. A breakdown in data migration due to faulty equipment or outdated products is something that should never occur if proper preparation takes place. Before the big move occurs, test everything, replace anything that is outdated or not working and then test to ensure that data is able to migrate properly. Lastly, when moving to a new data center, it is important to ensure that the new center can meet all heating and cooling needs and, most importantly, capacity needs with room to grow. While energy efficiency and lower costs of operation are important, the last thing you want when moving to a new data center is to have to move again in a few years because the needs of a data center were not properly anticipated. Leave room to grow so that a data center move is not just a short term solution but a long term solution for your data center.
Posted in computer room construction, Computer Room Design, Construction Industry, Data Center Build, Data Center Construction, Data Center Design, data center equipment, Data Center Infrastructure Management, data center maintenance, Datacenter Design, DCIM
Tagged computer room construction, Data Center Construction, data center design, data center equipment
To do an IT system upgrade or not to do an IT system upgrade, that is the question. Except, it is not really a question because, for data centers, IT upgrades are unavoidable. Any time an IT system upgrade is required, whether small or large, it will inevitably require more energy usage. This means that every time an IT upgrade occurs, operational costs will most likely increase. With so much focus on being energy efficient, and with the rising costs of energy usage, many data centers are trying to find ways to avoid increasing operational costs through improved energy efficiency while still maintaining the capability to upgrade their IT system as needed. Data Center Knowledge points out statistics about just how much energy data centers are using and why every IT system upgrade and increased energy usage is a concern for many, “Although the dire predictions of the 2007 EPA report on data center energy consumption have not panned out, there are still ongoing energy consumption concerns and data centers are not off the hook. Earlier this year, a report by Greenpeace criticized big data centers for using dirty energy (coal, gas, nuclear) as opposed to clean energy (wind, solar). A more recent report from the Natural Resources Defense Council (NRDC) claims waste and inefficiency in U.S. data centers – that consumed a massive 91 bn kWh of electricity in 2013 – will increase to 140 bn kWh by 2020, the equivalent of 50 large (500 megawatt) power plants. However, the 2014 Uptime Institute annual data center survey reveals that data center power usage efficiency (PUE) metrics have plateaued at around 1.7 after several years of steady improvement.”
With IT system upgrades that bring increased energy consumption as well as increased operational expenses, an evaluation of overall energy usage must be completed. You cannot endlessly upgrade and grow without evaluating and consolidating over time. By consolidating energy usage through a variety of other means you can lessen the impact of an IT system upgrade on the bottom line. Whether improving your heating/cooling system to improve energy efficiency, or consolidating servers, racks or more you can reduce energy usage in other areas within the data center to accommodate IT system upgrades. The best way to ensure that you are maintaining balance and, hopefully, constantly improving, is to routinely evaluate needs so that there is not wasted energy for unused equipment or other factors. A data center will run much more efficiently with a data center manager and team that understands just how much data is truly being used, as well as where energy is being consumed in large amounts within the data center. With a well rounded understanding, and routine evaluations of data center energy usage, IT system upgrades can occur without drastically impacting a data center’s overall energy usage or operational costs.
Just like a home or an office, a data center must be cleaned. Cleaning is not just to make it visually appealing, but to keep a data center running properly and efficiently. Data center cleaning may not sound like a high priority on the list of important things to consider with a data center. Most people probably think about things like rack density, preventing downtime, and heating and cooling issues. But cleaning a data center properly and routinely is incredibly important. When a site has visible dirt, or chemical odors, it is a major concern because it could mean there is chemical build up or problematic dust that is leading to the corrosion of important components which means damage to critical equipment. This is not only frustrating but expensive on many levels.
The best way to ensure your data center stays clean is to limit and control what comes in from the outside. Only a true “clean room” needs things like shoe covers but that does not mean that other rooms in a data center should be neglected just because they have not been specified as “clean rooms.” A data center should be cleaned only by professional data center cleaners. The reason for this is that there are many critical things, both large and small, that could easily be damaged or compromised in some way if an inexperienced cleaner is used. A data center cleaner will be cleaning everything including things like floors, subfloors, cabinet racks, server racks, walls, ceilings and more. In a data center more than just dust accumulates. Chemical residue tends to accumulate as well. This could be seen or unseen which is why it is important to use a professional. Even microscopic dust or chemicals must be removed to ensure a clean data center. Dust and particulates, when left to accumulate or simply unattended to, can interfere with critical functions and lead to data loss and media errors. Particulates can settle or move, migrating to different areas of the data center to cause major problems. If they make their way into air ducts they will simply be recycled into the room and continue on their path of destruction. Any time this happens, critical information could be lost and networks could experience downtime. As we all know, downtime can be incredibly costly, even if it only lasts a matter of seconds so regular cleaning is very important to prevent this from happening. In addition to having a yearly deep cleaning and regular maintenance cleaning, insist that those entering the data center wear clean clothes or cover ups so that you can best protect your data center.
Loss of Power
Data center downtime is one of the biggest fears of any data center manager. Data centers can experience downtime for a number of reasons, some within control and some completely outside of control. For instance, any poor weather ranging from a heavy storm to a natural disaster could potentially knock the power out in the area in which the data center resides. No matter what the source of downtime is in a data center, one thing is certain, data centers must have a backup in plan. But, this is not news. Every good data center has backup. The next question is – is that enough? The answer is a resounding no. To ensure that data center information and functionality is protected it is critical that there is a backup for the backup in place. Is this redundant? Yes. But ultimately, it could save money, jobs and ultimately, the entire business altogether. Data Center Knowledge points out just how costly downtime is for a data center and, by extension, a business, “Unplanned data center outages are expensive, and the cost of downtime is rising, according to a new study. The average cost per minute of unplanned downtime is now $7,900, up a staggering 41 percent from $5,600 per minute in 2010, according to a survey from the Ponemon Institute, which was sponsored by Emerson Network Power. The two organizations first partnered in 2010 to calculate costs associated with downtime.”
When you evaluate a data center, its power and capacity needs, and endeavor to create true redundancy in backup power supply you have to ensure that the redundancy is continuous. If it is not, major problems could arise. When creating the right system for your data center you have to try to anticipate future growth and needs. What worked yesterday for data centers no longer works for today. It is a continuously changing world and, for this reason, it is also important to do regular audits of your backup system to ensure true redundancy has been achieved and is still functioning as you move forward. The key power components of a data center include the backup generator, uninterruptible power supply (UPS), internal power supplies, power distribution unit (PDU) and much more. A fully redundant power supply will have adequate amounts of power supply to completely support the data center and all of its components with no single points of failure. Should a power outage occur, a data center will remain completely functional with this redundant backup power system in place. Not all data centers need this elaborate of a backup for their backup in place but, if the environment is running a mission critical project this type of backup is absolutely necessary. Backing up the backup may seem redundant, and it is, to save data centers from frustration, loss of time and significant loss of money.
With an influx of modular data centers it may leave many wondering if traditional legacy data centers are a thing of the past. Traditional data centers are likely not going away completely but there are certain applications in which modular data centers are a more ideal fit. When designing a data center there are many “future proof” considerations that must be factored in to the design process so that a data center does not become unusable or outgrown. But, these same considerations often lead to traditional data centers being inefficient. Because of this, a decision has to be made as to what is best – a traditional data center or a modular data center.
In the design process of a traditional data center there is often added room and extra heating and cooling capacity to accommodate growth and future needs. This methodology can help a traditional data center adapt and grow but it can also lead to an inefficient use of space and higher energy bills. Within a traditional data center energy efficiency can be achieved by taking advantage of strategies like hot/cold aisles, ceiling ducted air containment and more. When these strategies are used, a traditional data center can be efficient while still having room to grow and serving the existing needs of the data center. The problem with traditional data centers is that, often, future growth is incorrectly anticipated and if growth occurs more rapidly than expected, a traditional data center can be quickly outgrown.
Modular data centers present a number of benefits, particularly for new data center builds. When starting from scratch, modular data centers allow you to start with current needs and can then be scaled up or down as future needs arise. Data Center Knowledge discusses the importance of scalability when designing a data center, “With a repeatable, standardized design, it is easy to match demand and scale infrastructure quickly. The only limitations on scale for a modular data center are the supporting infrastructure at the data center site and available land. Another characteristic of scalability is the flexibility it grants by having modules that can be easily replaced when obsolete or if updated technology is needed. This means organizations can forecast technological changes very few months in advance. So, a cloud data center solution doesn’t have to take years to plan out.” For businesses that do not have the space or budget for a full-blown data center a modular data center is an ideal solution. Modular data centers streamline the process and make it easy to quickly and efficiently get a data center up and running. They often work with minimal space but maximum efficiency to quickly and easily launch a data center. When determining whether to design a traditional data center or a modular data center space, budget, rack density and more must be considered. Discuss these things with a data center contractor to determine whether a traditional data center or a modular data center is the best fit for your unique needs.
Data center consolidation projects, and particularly federal data center consolidation projects, are taking place all over the country. Data center consolidation projects can prove immensely beneficial and often saves a significant amount of money in the long run. Often, outdated data centers are not energy efficient and prove to be very wasteful. By consolidating old data centers and moving over some equipment and all of the data to newer data centers energy consumptions can be greatly improved. A significant amount of money can be saved through consolidation and the U.S. Government Accountability Office shows just how much through reporting recent statistics regarding federal data center consolidation projects, “Of the 24 agencies participating in the Federal Data Center Consolidation Initiative, 19 agencies collectively reported achieving an estimated $1.1 billion in cost savings and avoidances between fiscal years 2011 and 2013. Notably, the Departments of Defense, Homeland Security, and Treasury accounted for approximately $850 million (or 74 percent) of the total. In addition, 21 agencies collectively reported planning an additional $2.1 billion in cost savings and avoidances by the end of fiscal year 2015, for a total of approximately $3.3 billion—an amount that is about $300 million higher than the Office of Management and Budget’s (OMB) original $3 billion goal. Between fiscal years 2011 and 2017, agencies reported planning a total of about $5.3 billion in cost savings and avoidances.” Many data center managers need to take a lot into account to ensure that the transition between data centers during consolidation is smooth to prevent downtime or overloading in the new data center. As data centers scramble to meet energy and data demands, data center managers have to have a comprehensive and accurate knowledge of how much power and space will be needed in a new data center so that it can truly meet the needs. Additionally, cooling needs need to be anticipated because if not, things can become overheated and energy efficiency diminished.
Truly understanding the capacity of an existing data center that is being consolidated and the power, data and cooling needs necessary to accommodate it is easier said than done for data center managers. Amongst the things that need to be understood and anticipated are current density, design rating, bulk power capacity, bulk cooling capacity, data center capacity, physical site details, equipment and more. Determining what equipment can be brought from an existing data center to a new one will help anticipate physical space needs. Additionally, it is important to understand what existing software licenses can be transferred and what may need to be cancelled or renewed. While these are all important, it is simply skimming the surface of what needs to be anticipated in a data center consolidation project. A new data center that is taking on all of this from a data center consolidation project needs to not only accommodate the addition but be capable of expansion and growth in the future. While unanticipated problems are inevitable in any consolidation project, preparedness is the key to minimizing downtime and optimizing uptime and efficiency.
Posted in Data Center Build, Data Center Construction, data center cooling, Data Center Design, data center equipment, Data Center Infrastructure Management, Datacenter Design, Power Management
Tagged Data Center, Data Center Construction, data center design, data center equipment
A data center often prides itself on its ability to maintain uptime in the face of significant power and data loads. Downtime is seen by customers and consumers as simply unacceptable and while there are many ways to avoid downtime, one important protection all data centers need to utilize is redundancy. Data center hardware and power needs are constantly evolving and because of this, it can be difficult to anticipate and properly ensure power supply redundancy. With a dual-path power supply true redundancy can be achieved but it is critical that both paths are capable and managing all equipment and power needs in the data center. ComputerWeekly describes the importance of redundancy in a data center, ” Redundancy has a negative connotation when the duplication is seen by the business as unnecessary. Yes, for some businesses this datacentre capacity excess is an issue, but for the majority, the other form of redundancy – the provisioning of a datacentre to survive a range of failure scenarios – has become even more of an issue. IT infrastructure is part of an organization’s DNA. If someone were to cut off the IT service for an organization, it would not be a small snag, but a corporate catastrophe for its operations. Business processes would halt, customers would be left stranded, suppliers would be unable to know what was required to be delivered, the organization would struggle to pay its employees what they are owed, communication and collaboration would be severely impaired. The overall availability of an IT platform means that an approach of a single application on a single physical server with dedicated storage and individual dedicated network connections is a strategy to oblivion. It is incumbent on IT to ensure that the IT platform can continue to operate through failures – as long as the cost of doing so meets the organization’s own cost/risk profile.”
The purpose of having a dual path is that, should a failure occur in the power supply, the other power supply will be able to pick up exactly where the first left off and maintain uptime. Having a redundant power supply today is simply best practice for any data center. One of the most important parts of creating true redundancy is ensuring that, should a failure occur somewhere in one path, it will not affect or impact the other path in any way. While many data centers choose to employ redundant power supplies, many data centers fail to realize just how much power their supplies will actually need for true protection. The sad fact is, many data center managers simply do not have an adequate grasp on the power supply needs of all of the equipment in the downstream. Because new equipment is constantly being added to data centers, the additions can frequently get overlooked and suddenly, without realizing it until it is too late, the power needs exceed the redundant power supply. The power demands should be constantly checked so that the power redundancy is never exceeded. The best way to do so is through the implementation of a monitoring system that routinely checks and ensures that all is as it should be. When discussing redundancy, many may raise an eyebrow at the concept and think it is not very efficient. But, to properly run a data center, redundancy is necessary. To ensure optimal efficiency, it is best to run the data center on the most efficient power supply and reserve the less efficient power supply as the redundant supply that is only used should a failure occur.