AIRFLOW MANAGEMENT FOR THE EFFICIENT MAINTENANCE OF A DATA CENTER

October 9, 2012 By

Since the inception of the computers in the early 60’s, airflow within the space has been an important consideration from the design perspective. But this aspect has often been misunderstood. Considering the low density of computing that was done

back in those years, this design factor did not make its presence felt. However, today with enormous amount of computing done and high performance servers doing double the amount of work compared to their predecessors, airflow management in a data center becomes important.

The most important principle behind an effective airflow management system is-

  • Minimizing the mixing of inlet and outlet exhaust gases: The purpose of this can be achieved by deploying the following-
  1. Cool air from the cooling units (CRAC) are allowed to reach only the server intakes.
  2. Hot air from the server goes only to the CRAC exhausts.
  • The second important principle to be borne in mind is that the CRAC temperatures from the returns are to be raised to high temperatures while compensating with lower temperatures for the inlet air.

This would in turn allow the servers and the heat rejection systems to operate efficiently. This also makes way for reduced operating costs with greater server accountability.

After the analysis of the principle, it is important to understand the maintenance of these systems on a regular basis. How does the operator know if the cooling systems are operating efficiently? For this purpose we may look into the raised floor data center, whose basics are synonymous with any data center.

Data center layout needs to be cross checked before going ahead with the installation of the cooling systems. The most efficient layout will have the Hot/Cold aisle configurations in racks with an ability to extract hot air from the racks and return to the CRAC exhaust. This means that the CRAC has to be placed perpendicular to the rows.

Rows ideally should be 8 or more racks long with a full 2 tile wide cold aisle. Racks fronts should be even with the tile edges so that any tile in the cold aisle can be removed or replaced to allow adjustment of cold air volume. One of the biggest issues with blanking panels is making sure they are replaced when work is done in racks. Racks should be 3 full tiles away from the CRACs to minimize low flow from tiles that are too close to CRACs. Where possible, place less server load in racks at the ends of the aisles and higher loads in racks towards the middle of the row. Additionally load racks from bottom to top, keeping the highest server loads lower in the racks.

Containment is becoming a popular strategy for airflow management in data centers. Its specialty is that it uses a physical barrier to reduce, or in some cases completely eliminate mixing which allows higher CRAC supply and return temperatures.

Every data center is different and has varied airflow challenges and opportunities but the subset of problems in each has known and proven solutions. Airflow management, properly applied, will reap benefits in energy efficiency, IT capacity and server reliability.

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.Please leave your views and comments on DCT Forum

 

 

Share on TwitterSubmit to StumbleUpon
Filed Under: Articles, Data Center Cooling, Data Center Design Tagged With: , , ,

DATA CENTER COLOCATION

October 8, 2012 By

Two of the most important aspects of a data center’s success is the availability and business continuity. In today’s growing economy, every enterprise feels the pressure from internal and external demands by the IT sectors. One needs to understand that it is not an easy task to predict the need of the hour and design data centers accordingly. When the key services of a data center, availability and business continuity are endangered, the existing data center cannot shoulder all commitments, it becomes necessary to upgrade the facility.

Every company faces design and location limitations at one point or the other. When an undesirable situation arises, in order to continue with the services, without compromising on the quality, data centers can be upgraded to levels necessary to provide required performance.

A colocation data center can resolve issues of economic crunch and location glitches. Colocation centers are those facilities which provide customers with space for their servers and bandwidth. They house all the hardware, equipments, servers, data storage devices of other firms and connect these to networks across the globe. It literally puts the server in the client’s hands. Colocation is completely different from the traditional hosting although it might be a little mind boggling for new entrants. Hosting providers own the entire hardware and software infrastructure; they only provide space for websites, and services that serve up an application. In contrast colocation customers own their servers and hardware; colocation providers ensure the security of these with efficient power and cooling backup.

Managed hosting limits the ability of a client to customize the server environment. Colocation offers complete freedom to customize the servers. The client purchases the equipments, installs the software, and configures it according to the needs. After the system has been configured it is installed in a space provided at the center. The colo provider ensures a safe IP address, electricity, and connection to the network.

There are more than 400 colocation service providers with a wide array of services and price points to match. Some colo providers (as they are normally called) categorize their customers and prefer to keep their options limited to a particular sector.

Clients who need redundant equipment should look for data center colocation. It also acts an alternative to load balancing solutions. For any client who experiences heavy traffic and usage on a single day, load balancing becomes an important criteria to prevent the server from crashing. Colocation centers provide the necessary space for an effective load balancing strategy. Load balancing goes hand in hand with redundancy, colocation is effective in this regard.

Location is another important criteria an enterprise needs to ensure. Since the clients are the owners of the servers, having a colocation facility close to the workplace is important. Carrier flexibility, when the client wants one separate dedicated carrier line, is also a very important area that one cannot ignore. For this purpose, quality carriers are available.

Summary

Colocation facilities are a growing trend in the present economy due to-

  • higher reliability due to redundant systems
  • full time monitoring by live humans
  • lower network latency and higher bandwidth at a lower cost
  • Specialist staff, like network and facilities engineers, which would not be cost effective for any single client to keep on the payroll.
  • Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.Please leave your views and comments on DCT Forum
Share on TwitterSubmit to StumbleUpon
Filed Under: Articles, Data Center Design, Data Center Solutions, Hosting / Colocation Tagged With: , ,

STRUCTURED CABLING

October 5, 2012 By

Structured cabling involves the connectivity of different smaller elements that specify the wiring for data centers, offices, and buildings using various kinds of cables adhering to specified standards. These standards define the layout of the cables according to the requirements of the clients and the data center as a whole.

Data centers and storage area networks are the fastest growing streams of technology in the IT sector. It has been reported by a recent study that the growth and retention requirements of these centers have gone up by 50% this year. This type of growth is governed by various legislative agreements on how much data is to be stored and for how long.

Cabling of a data center needs to adhere to different standards of specification. Amongst them the most important standard that one needs to know for structured cabling is the TIA standard.

The TIA standard specifies the minimization of design and management of structured cabling within data centers.

Before analyzing the TIA standard, it is important that we understand the need for structured cabling for a data center.

 

Need for structured cabling

Most of the time data centers and SANs are constructed without having considered the implications of frequent additions, moves, and expansions. Some systems like the computers and single physical servers are normally installed by the company’s own technicians and crew. This crew is competent when it comes to the perspective of their own equipments. But the data centers house in them varied, disparate equipments, and data storage devices. Using such practices inevitably causes inefficient management of critical conditions. Critical conditions could range from advancement in technology to the addition of new products and services.

In the early years, a wide variety of cabling and architecture were common, but difficult to manage. This form of undesirable situation lead to the formation of the TIA/EIA-568 Commercial Building Cabling Standard, which eventually changed the way cabling for commercial buildings, telecommunication sectors, and data centers were done. This introduced a new and modernized way of cabling that is the ‘structured cabling’.

 

Many cable and copper industries developed new connectivity products that dramatically offered advantages to the data centers and SANs. Most of the experts and their successors, who developed the TIA-568 and its next called the TIA-568a, developed another standard called the SCS.

The TIA-568 was designed to suit the requirements of commercial buildings.

The TIA-942, Telecommunication Infrastructure Standard for data centers is bound to have immense effect on the data center and SAN as profoundly as the TIA 568 has on commercial buildings.

This new standard allows the SCS concepts to be implemented in the disparate equipments very early in the building design process. This particular standard views the whole data center as an integrated system with smaller ancillary elements. As a result, it interlinks many components like location, access along with architecture and electrical components to a most important concept of redundancy.

The TIA 942 includes seven spaces and two cabling subsystems within the data center.

  • Ø Seven spaces include-
  • Computer room
  • Telecommunications room
  • Entrance room
  • Main Distribution Area(MDA)
  • Horizontal Distribution Area( HDA)
  • Zone Distribution Area( ZDA)
  • Equipment Distribution Area (EDA)

 

The cabling subsystems in TIA 942 include the horizontal and backbone.

 

The first five spaces generally involve many connections like high density panels and racks using fiber connectors like LC. The entrance room is the interface within the campus and is similar to the entrance room of a commercial building. MDA is the area where the main cross connect is located. HDA houses the horizontal cross connect. ZDA is an optional space and is where the zone outlet is located. EDA is where the cabinets, servers, and racks are located. It is similar to the working area of a commercial building.

 

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.

Please leave your views and comments on DCT Forum

Share on TwitterSubmit to StumbleUpon
Filed Under: Articles, Data Center Design, Data Center Equipments, Fiber Optic Tagged With: ,

CLOUD COMPUTING AND VIRTUALIZATION

October 4, 2012 By

In the present scenario of changing technology, the IT sector is keen on reducing costs, and improving performance. Innovative use of the advanced techniques is a logical solution, but is very elusive. Data centers need to optimize infrastructure and deploy strategies to improve data center operations. These would enable centers to leverage the benefits of cloud computing and virtualization for their economic benefits.

Cloud computing and virtualization are two relatively similar ideas, but with a whole new horizon. The whole process of implementing the infrastructure requires a lot of detailed discussion about the type of data center and its requirements.

Cloud computing is a scalable platform that is used for the delivery of a service or a capability, to build and run applications over the network. It refers to the IT services that are deconstructed from the underlying infrastructure, and provisioned when needed as well as providing infinite resources to meet the growing demands. It entrusts remote services to the user’s data via computing and networking. The user’s data is stored in a remote location, whereas the end client can access the application through a web browser or a simple mobile app.

Cloud computing fundamentally changes the way IT services are delivered to clients. IT sectors need not concentrate on developing a separate hardware and software encrusted infrastructure for all their servers and storage devices. This would mean that the companies also need to upgrade their servers based on the changing trends. This facility of cloud computing requires neither ownership nor installation of dedicated resources. Cloud computing offers a wide array of benefits:

Reduced cost: Reduction in the cost is definitely an added advantage to any data center operative. The organization would have to invest only in the required capacity and not on unnecessary infrastructure and maximum capacity.

Flexibility: A data center always needs to forecast its future requirements. It needs to be well equipped to situations of migration and need of excess workload. Cloud computing eliminates these undesirable consequences.

Location independence: As cloud computing does not use a physical infrastructure, users can access any application irrespective of their location or the device they are using.

Scalability: A major problem that data centers incur is the expansion plan. Cloud computing offers “on demand” services without having to look for onsite resources.

Security: Access to the log audits is very difficult, hence ensuring security to the data stored. Since the data is distributed over a widespread area and relying on multi-tenants, concern over data loss is relatively minimized.

Maintenance: Maintenance of cloud computing applications is simpler as it does not involve installing applications on every user’s device. It can be accessed from any location.

Server virtualization refers to the servers and data storage devices that are simulated using software. It is the division or segregation of servers. Every single virtual server runs on its own and as a part of another physical server. Many virtual servers can run on a single physical server. These virtual servers can be isolated and can be handled singly, even away from the physical servers.

Through virtualization, you can install a software only once and be rest assured that everyone will have access to it. You don’t need multiple licences to make the software available to all your employees. Since you are technically installing it only on one system, you are not violating any laws either. Same is true with storage. This technique avoids the need for data replication, thus saving storage space.

Virtualization can bring higher utilization, better security and easier management.

Server utilization can definitely bring in some great amount of flexibility and energy efficiency into data centers.

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.

Please leave your views and comments on DCT Forum

Share on TwitterSubmit to StumbleUpon
Filed Under: Articles, Cloud Computing, Data Center Design, Data Center Solutions, Virtualization Tagged With: ,

DATA CENTER CABLING REQUIREMENTS

October 3, 2012 By

Data centers are not just facilities that house a large number of servers and data storage devices. The whole data center works as a system, with all the entities working together to protect and maintain all the equipments, data transmitted, and the numerous connections between the server and the areas outside the data center. All these entities are interrelated and are connected via a structured cabling system. Designers and engineers are working towards achieving a suitable cabling system, which would define the standards of flexibility, scalability, operational costs, and installation overheads. Designing a cabling system is a complex issue and needs to be weighed considerably before installation.

The factors that need to be considered for an effective cabling system are-

  • Scalability: The life expectancy of the cabling infrastructure needs to be calculated between the different components. The future requirement of the center needs to be visualized. It has to be ensured that there is ample space for future additions to servers, ports, and switches.
  • Flexibility and Manageability: It has to be clarified if the center needs often moves, additions, and changes. Anticipation of the need to reconfigure the functional areas becomes very important. Ensure there is ample space for extra cabling and router installation.
  • Availability: Determine the required uptime necessary for establishment, power, and cooling. Also determine the level of security for each functional area, and the data center as a whole.
  • Total cost of cabling: Calculate the investment that goes into the infrastructure of cabling by taking into consideration the scalability, flexibility, and availability requirements.

Each of these factors play very important role in defining the design of a cabling system.

The design of the cabling system has to abide to certain set international standards and regulatory authorities such as-

  • TIA 942: Telecommunications Infrastructure Standard, April 2005 – American
  • EN 50173-5: Information technology – Data Centers

Smaller scope, European perspective

  • ISO/IEC Draft 24764: Information technology

      Global Generic Cabling for Data Centers

Industry cabling standards are designed to protect the end-user. They provide a firm foundation and guidelines for maintaining high levels of cable performance. Vendors design the cables according to the prescribed standards and topololgies, so that the installation as per company standards does not have compatibility issues. These standards are reviewed and revised every five to ten years, to keep in pace with the changing technologies and advancements. So all it needs is a bit of understanding the standards, and implementing it as per requirements.

Practices in maintaining the cabling system-

  • Avoid bundling up of cables, and stocking up multiple cables one on top of the other. This would degrade the performance of the cables. Do not pile copper and fiber cables together, as heavy copper cables crush the vulnerable fiber cables placed underneath.
  • Ensure that the cabling is compliant to the industry standards.
  • Applying stress on the cables should be avoided.
  • Extra tension, and pressure above permissible limits should be avoided.
  • Avoid clumsy infrastructure, do not add many components resulting in a crowded location.

Choose the strongest foundation for the cabling support, achieving future goals and requirements. Since each environment is different, it is difficult to choose the correct cabling system. Following the guidelines and standards, will help in the effective employment of the cable infrastructure.

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.

Please leave your views and comments on DCT Forum

Share on TwitterSubmit to StumbleUpon
Filed Under: Articles, Data Center Design, Networking / Security Tagged With: ,

DATA CENTER VIRTUALIZATION

October 1, 2012 By

Server virtualization refers to the servers and data storage devices that are simulated using software. It is the division or segregation of servers. Every single virtual server runs on its own and as a part of another physical server. Many virtual servers can run on a single physical server. These virtual servers can be isolated and can be handled singly, even away from the physical servers.

At a business and enterprise level, virtualization of data centers helps in reducing equipment costs, improves quality and system availability.

All the resources need to be managed well like the software and hardware in order to create virtualization efficiently and in an affordable manner.

In the traditional and conventional methods, every new application required a separate server. This inturn adds overhead costs and takes several weeks to be employed. This means that there are huge stacks of servers staged in different rooms. This would lead to excess power consumption and energy wastage.

Virtualisation is a technique of creating a virtual pool of servers, operating systems, storage devices and network resources. It enables a single user to access multiple physical devices at the same time. With this, one operating system can control the operation of multiple computers or vice versa.

Through virtualization, you can install a software only once and be rest assured that everyone will have access to it. You don’t need multiple licences to make the software available to all your employees. Since you are technically installing it only on one system, you are not violating any laws either. Same is true with storage. This technique avoids the need for data replication, thus saving storage space.

The drawback with data centers is the challenge you will face while scaling up as and when the need arises. Your data center host must have rack space to accommodate an extra server or two and also must be equipped to handle an increase in cooling and power needs. Of course, there is a problem of your resources going on standby mode when not in use, too.

Storage virtualization is one type of virtualization that greatly helps in reducing the operational costs and improving the agility of performance. It means that the large number of servers and storage devices are virtually stacked up in a pool, for e.g. a virtual storage pool. Storage virtualization leads to the next generation of data centers.

Computing virtualization is also growing very fast. Some of the leading companies in the space have developed cross-platform virtualization software that allows data center managers to cluster and partition servers. .With the help of these tools, many virtual servers can be offered on an a single server or a single server can offer the roles of many different virtual servers.

Networking virtualization can occur at the LAN or the WAN. On the LAN, virtual LAN technology allows network managers to isolate and partition LANs running on a single physical LAN.

On the WAN side, the same effect can be achieved with MPLS, a technology that offers partitioning of WANs into multiple isolated virtual WANs running on a single physical infrastructure. Virtualization can bring higher utilization, better security and easier management.

Server utilization can definitely bring in some great amount of flexibility and energy efficiency into data centers.

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.

Please leave your views and comments on DCT Forum

Share on TwitterSubmit to StumbleUpon
Filed Under: Articles, Data Center Design, Storage / Server, Virtualization Tagged With: , ,

AIRFLOW MANAGEMENT

September 28, 2012 By

Data centers consume about 25 to 30 times more electricity compared to the normal office spaces. This calls for an energy efficient design of the data center that can save money and reduce electricity use.

 Data center design is a relatively new stream that contains a dynamic and evolving technology. The most efficient data centers incorporate modern design technologies that are cost-effective and energy efficient. Short design cycles lead to incomplete assessment of the full design requirements. Most of these short design cycles ultimately lead to just scaling up the older versions of design of office spaces.

Modern data centers house server racks in a fashion that leads to concentration of the heat loads. In facilities of different sizes, starting from a small data center for a few office buildings to a large co-location facility; design of the center to precisely control air flow is of utmost importance. Air flow through the room for efficient removal of the accumulated equipment heat has a strong impact on the reliability and the energy efficiency of the entire data center.

Air management includes all the minute details of design that are required to minimize or curtail the mixing of cool air supplied to the room with the hot air rejected from the room. When it is designed correctly, it helps in reducing the operating and maintenance costs of the equipment and other issues caused by the thermal heating of the devices.

The main design issues related to air management are:

  • Location of the data center
  • Location of the cooling equipments
  • Equipments required for intake and exhaust ports
  • Configuration
  • Air flow patterns in the room

Principles of air flow management:

  • Use of hot and cold-aisle configurations can double-up the cooling efficiency of the data center.
  • With the aide of an airside economizer, air management can reduce data center cooling costs by over 60%
  • Removing hot air immediately through the exhaust improves efficiency rather than mixing the hot air with the incoming cold air.
  • Equipment environmental temperature specifications refer primarily to the air being drawn in to cool the system.
  • A higher difference between the return air and supply air temperatures increases the maximum load density possible in the space and can help reduce the size of the cooling equipment required, particularly when lower-cost mass produced package air handling  units are used.
  • Poor airflow management will reduce both the efficiency and capacity of computer room cooling equipment.

 Therefore an effective air flow management system can bring down the costs and power consumption rates to a great extent. Design of these systems also play a very important role in the quality, reliability, and security of data centers.

Share on TwitterSubmit to StumbleUpon
Filed Under: Articles, Data Center Cooling, Data Center Design Tagged With: ,

ComEd Launches Data Center Energy Efficiency Program

August 9, 2012 By

Building on the assets that have made Chicagoland and northern Illinois a top market for data centers, ComEd has launched a significant new energy efficiency program that will spur regional business development in the digital economy sector.

Through this new incentive offering,  part of the Smart Ideas for Your Business® program, ComEd will work with data center designers, owners, and tenants to strengthen operations and maximize energy efficiency.  By providing per kWh financial incentives and technical assistance, ComEd will help these businesses lower energy use and save money on electricity bills. Whether businesses are building new facilities, renovating or expanding, ComEd will work with them to achieve a permanent reduction in electrical usage.

Data centers are the large facilities used to house computer servers and other systems that power our modern, digital economy.  They also are major consumers of energy, requiring a great deal of electricity to both power and cool equipment. They can consume 100-200 times more electricity than standard office spaces.

“The reliance on digital technology to support virtually everything we do is increasing on an almost daily basis and businesses count on uninterrupted power to support all of that technology,” said Val Jensen, ComEd’s senior vice president of Customer Operations.   “ComEd’s reliability performance is already in the top third nationally and is on target to become even stronger due to investments we are making under the Energy Infrastructure Modernization Act (EIMA), so we offer data center customers the reliability they need and technical and financial support to run an energy efficient operation.”

ComEd has two dedicated account managers who are specialists within the data center industry.  They currently serve as energy liaisons to over 60 data centers in northern Illinois. These experts will work with both new and existing businesses to reduce utility bills by helping to offset the cost of building or modernizing their equipment and facilities.

The combination of  a skilled workforce, a strong fiber optic infrastructure, a large peering network among fiber optic providers and high financial and enterprise server demand from a large base of Fortune 500 companies and financial businesses, including 28 located in the metropolitan area, have contributed to the Chicago area’s popularity as a location for data centers.

In addition, for companies seeking reliable and affordable power options, the region offers a robust competitive electricity market with more than 30 suppliers; extensive energy efficiency programs; low-carbon generation sources from the highest concentration of nuclear power plants in the country and growing renewable resources, including wind and solar power.

ComEd’s new incentive offering as part of the Smart Ideas for Your Business program builds on that strong foundation. Last year, energy efficiency projects were completed at several customer sites. Latisys, a leading provider of IT infrastructure services from colocation to cloud, implemented advanced cooling systems in their Chicago data center.  By better managing air flow, they can conserve energy and lower cooling costs.

“The incentive from ComEd enabled us to implement hot aisle and cold aisle containment systems in our Tier III data center,” said Nabon Marsico, Director of Data Center Operations, Latisys. “This is already translating to significant savings while allowing us to provide more concentrated and precise cooling for our hosting and colocation customers.”

ComEd’s new program is part of the Smart Ideas for Your Business portfolio, which recently entered its fifth year.  The Smart Ideas program encourages customers to save money and reduce energy consumption by helping them offset the cost of energy efficiency measures.

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.

Please leave your views and comments on DCT Forum

Share on TwitterSubmit to StumbleUpon
Filed Under: Data Center Design

How to Calculate Data Center Cooling Requirements

July 16, 2012 By

Electrical equipments when operated over long hours produce heat, which must be removed to prevent the equipment from getting damaged. Sizing a cooling system requires an understanding of the amount of heat produced by the equipment contained along with the heat produced by the other heat sources.

The total heat output of a system is the sum of the heat outputs of the individual components. The complete system includes the UPS, Power Distribution, Air Conditioning Units, Lighting, and People. Fortunately, the heat output rates of these devices can be easily determined through simple and standardized rules.

The heat output of UPS and Power Distribution systems consists of a fixed loss and a loss proportional to operating power. Heat generated by lighting and people can also be readily estimated using standard values. The only information needed to determine the cooling load for the complete system are a few readily available values, such as the floor area in square feet, and the rated electrical system power.

A quick estimate using simple rules gives results that are within the typical margin of error of the more complicated analysis. The quick estimate also has the advantage that it can be performed by anyone without specialized knowledge or training.

The prior analysis ignores sources of environmental heat such as sunlight through windows and heat conducted in from outside walls. Many small data centers and do not have walls or windows to the outside, so there is no error resulting from this assumption. However, for large data centers with walls or a roof exposed to the outdoors, additional heat enters the data center which must be removed by the air conditioning system.

If the data center room is located within the confines of an air-conditioned facility, the other heat sources may be ignored. If the data center has significant wall or ceiling exposure to the outside, then a HVAC consultant will need to assess the maximum thermal load and it must be added to the thermal requirement of the complete system determined in the previous section.

The determination of cooling requirements for IT systems can be reduced to a simple process that can be done by anyone without special training. Expressing all measures of power and cooling in Watts simplifies the process.

For larger data centers, the cooling requirements alone are typically not sufficient to select an air conditioner. Typically, the effects of other heat sources such as walls and roof, along with recirculation, are significant and must be examined for a particular installation. The design of the air handling duct work or raised floor has a significant effect on the overall system performance, and also greatly affects the uniformity of temperature within the data center. The adoption of simple, standardized and modular air distribution system architecture, combined with the simple heat load estimation method could significantly reduce the engineering requirements for data center design.

 

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.

Please leave your views and comments on DCT Forum

Share on TwitterSubmit to StumbleUpon
Filed Under: Data Center Cooling, Data Center Design, Data Center Management Tagged With: , ,

How Not To Plan a Data Center

July 13, 2012 By

Lessons learnt from the NGA’s technology center at Belvoir

One of the floors of the National Geospatial-Intelligence Agency’s new Technology Center has been vacant for the past few years. The project is getting rebooted with $54.6 million funding, but it’s anyone’s guess how it will end up.

 

The NGA’s Technology Center located at Fort Belvoir is the subject of a Government Accountability Office (GAO) report tied to the Pentagon’s ongoing Base Realignment and Closure efforts. As part of the consolidation, the NGA in 2006 began constructing several new buildings its “Campus East,” including a sprawling, 2.4 million-square-foot headquarters and the four-story, $155 million Technology Center.

 

The tech center blueprints originally called for 108,000 square feet of data center space. Within 12 months, that space requirement was cut in half, to 55,000 square feet; taking into account the efficiencies of consolidating systems on more densely configured servers. The idea was to establish 22,000 square feet of data center space on the fourth floor, and do the same on the third floor.

 

Construction started in October 2007, but within a few months the IT contractor on the job, General Dynamics, contemplated that the third floor wouldn’t be needed to accommodate the sought-after 10 petabytes of storage capacity. General Dynamics explained that advances in server and storage technology made it possible to squeeze data into a smaller footprint.

With construction already underway, it was too late to omit an entire floor, and the facility was completed as planned in 2009.

 

The project has been plagued with problems from the start. When the NGA, whose intelligence work is used by the U.S. military, determined that it no longer needed the third floor of the Technology Center, it failed to notify the U.S. Army and the official project manager. The Secretary of Defense and Congress were left out of the loop, too.
The project’s shoddy documentation didn’t help. Department of Ocean Development  must provide a support of construction work, but the NGA didn’t mention the Technology Center in its February 2006 filing for the Fort Belvoir campus according to the GAO report. That part of the project was still in the “conceptual design phase,” the agency later explained.
The origins of the project go back to 2005, when the NGA set out to build its Campus East to serve as a new home for 8,500 employees, as its other offices were being closed. Each year, the NGA drafted a business plan for the project, but it failed to make any mention of its decision to mothball the third floor of the tech center. It wasn’t until last year that officials with the Army’s BRAC office, on a tour of the campus complex, realized that office space the size of a football field was empty.

 

There two important lessons learnt from the by the project management team are, that pristine data center space can almost always be put to good use if not now, then later. The second: When estimating data storage requirements, take your first estimate and multiply them by 10 folds.

Data Center Talk updates its resources everyday. Visit us to know of the latest technology and standards from the data center world.

Please leave your views and comments on DCT Forum

Share on TwitterSubmit to StumbleUpon
Filed Under: Data Center Design, Data Center Management Tagged With: , ,
« Older Posts
Newer Posts »