Friday, June 08, 2007

Recommended Reading: “The New Data Center”

There’s an interesting piece up at NetworkWorld.com on the leading trends in data center storage, titled “The New Style of Storage.” It touches on a variety of topics including e-discovery, eco-friendly storage technology, and virtualization.

This is part 3 in a six-part series that examines the latest technologies and practices for building “the New Data Center.” Taken together, it’s a bit of a long-read, but well worth the time. Check it out when you have a chance.

Be sure to take a look at parts 1 and 2, as well:
Part 1, The New Data Center –
Trends, Products & Practices for Next-Gen IT Infrastructure
Part 2, Defending Your Net – Tools and Tactics for Enterprise IT Security

Want to weigh in on what you’ve read? I’d love to hear it. Post your thoughts on the comments page for this entry.

Tuesday, May 15, 2007

High Density Devices Strain Data Center Resources

A few weeks back I commented on the current boom in data center development. Spurring this trend is the growing need for greater processing power and increased data storage capacity, as well as new Federal regulations which call for better handling and storage of data.

In the scramble to keep up with these demands, the deployment of high density devices and blade servers has become an attractive option for many data center managers. However, a new report from the Aperture Research Institute indicates that “many facilities are not able to handle the associated demand for power and cooling.”

The study, based on interviews with more than 100 data center professionals representing a broad spectrum of industries, reveals that the deployment of high density equipment is creating unforeseen challenges within many data centers.

Highlights of the report include:

  • While the majority of data center managers are currently running blade servers in their facilities, traditional servers still comprise the bulk of new server purchases. Mixing blade and non-blade servers in such small quantities can unnecessarily complicate the data center environment and make maintenance more difficult.
  • The rising power density of racks makes them more expensive to operate and more difficult to cool. More than one-third of the respondents said their average power density per rack was over 7KW, a scenario that setting the facilities up for potential data center cooling issues and unexpected downtime.
  • Respondents report that the majority of data center outages were caused by human error and improper failover.
  • What’s really jaw-dropping is that while more than 22% of outages were due to overheating, 21% of respondents admit that they don’t know the maximum power density of their racks. The report points out that “[o]ver 8% of respondents are therefore using high-density devices without tracking power density in a rack, dramatically increasing the potential for outages.”

High density equipment can help data centers keep up with business demands, but only if you can keep things running smoothly. Proper management of power and cooling is essential for meeting the end user's availability expectations. For more information on the various cooling challenges posed by high density rack systems, please visit our Data Center Cooling Challenges page at PTSDCS.com.

Tuesday, April 24, 2007

NEW from PTS: Computational Fluid Dynamic (CFD) Services

Behind the scenes at PTS Data Center Solutions, we’re always working to enhance our products, services and solutions in order to provide our clients with designs that offer optimum manageability and performance.

Our newest consulting service utilizes powerful 3-D Computational Fluid Dynamic (CFD) software to facilitate the design, operational analysis and maintenance of our clients’ data centers and computer rooms.

Here is an overview of the multiple applications of our CFD Services:

  • CFD Modeling as a Design Tool

By building CFD models of a mission critical space, engineers can quickly and efficiently review multiple design options. This allows for early detection of potential problems with air flow and heat distribution, thus permitting designers to provide an optimum solution.

  • CFD Operational Baseline Service

After the data center’s IT infrastructure has been populated, PTS uses CFD modeling to map the site and analyze the data center cooling characteristics down to the equipment level. By doing so, we can determine how variations in the position and design of equipment, as well as other factors, affect the room’s cooling profile.

  • Maintaining a CFD Modeled Computer Room

To ensure the high performance and manageability of a mission critical site, it is important to understand the effect that equipment changes will have before implementation takes place. Through CFD visualization, simulation and analysis, PTS’s consulting team can predict the impact of operational changes on the temperatures in the room. From there PTS is able to make recommendations for avoiding potential problems while planning for future growth. As part of the CFD modeling process, PTS maintains a complete asset inventory log as well as a detailed change order log, ensuring that infrastructure changes are tracked correctly.

If you’re interested in learning more about this data center consulting service, please visit our Computational Fluid Dynamic (CFD) Services page.

Request a Quote

To request a quote for PTS's CFD Baseline and/or Maintenance Services, please send an email to CFD@PTSdcs.com with the following information:

  1. The physical address of the location
  2. The square footage of the computer room to be modeled
  3. The number of server cabinets, racks, and stand-alone pieces of equipment in the computer room
  4. The number of IT infrastructure devices (servers, switches, routers, storage arrays, etc.) the computer room supports

Wednesday, April 11, 2007

Keeping It Clean in the Data Center

Spring is here. It’s the time of year when people throw open the windows, pull out the dust rags and fire up their vacuums for a burst of Spring Cleaning. This annual household ritual serves as a good reminder of the importance of regular cleanings within the data center environment.

Regularly scheduled site cleanings help to keep the data center environment free of dust, dirt and other particulates that can harm your operating systems and create health risks for employees. Particulates circulating within a data center can accumulate and interfere with electronics causing a variety of potential problems, including media errors and data loss.

A good rule of thumb is to schedule data center cleanings on a quarterly basis, or when particulate counts exceed the standards set by ISO 14644-8 or ISO 14644-9. By sticking to this cleaning routine, companies optimize the performance of data center equipment while cutting down on the cost of repairs. When you compare the cost of regular cleaning sessions to the overall financial investment in your data center, it’s a smart buy.

Choosing a Data Center Cleaning Service

Don’t grab a broom and dustpan just yet. While it’s good to clean both houses and data centers on a regular basis, that’s where most of the similarities end. Cleaning a data center is a delicate process that requires the services of highly-trained professionals who know how to safely handle mission critical equipment.

To help you select the right cleaning service, here are some tips:

  • Check the company’s references. In addition to the quality of the service, you want to make sure the company has experience dealing with facilities that are similar to your own.
  • Makes sure the company is insured for damages caused during the cleaning process. If an accident occurs, are you protected?
  • Evaluate the experience and training of the cleaning crew. For instance, are they trained to provide services per the requirements of International Standard ISO 14644?
  • Review the company’s cleaning methods to see if they use HEPA filtration vacuums and chemicals that are safe for use with electronics systems.
  • Be clear about your expectations for the service and establish parameters for cleaning. Will the technicians move equipment? Will they clean the sub-floor or above each rack? Are certain areas off-limits? What’s included in the service?
  • Look for a cleaning service that offers availability that meets your needs. In addition to yearly cleanings, will they be available for daily maintenance activities or in the event of an emergency?

Friday, April 06, 2007

Data Center Cooling: Approaches to Avoid

Data center cooling problems can compromise availability and increase costs. The ideal data center cooling system requires an adaptable, highly-available, maintainable, manageable, and cost effective design.

When working to design an effective data center cooling system, there are a number of commonly deployed data center cooling techniques that should not be implemented. They are:

  • Reducing the CRAC supply air temperature to compensate for hot spots
  • Using cabinet and/or enclosures with either roof-mounted fans and/or under-cabinet floor cut-outs, without internal baffles
  • Isolating high-density RLUs

Reducing CRAC Temperatures

Simply making the air colder will not solve a data center cooling problem. The root of the problem is either a lack of cold air volume to the equipment inlet or it is lack of sufficient hot return air removal from the outlet of the equipment. All things equal, any piece of equipment with internal fans will cool it self. Typically, equipment manufactures do not even specify an inlet temperature. They usually provide only a percentage of clear space the front and rear of the equipment must be maintained to ensure adequate convection.

Roof-mounted cabinet fans

CFD analysis conclusively proves that roof-mounted fans and under-cabinet air cut-outs will not sufficiently cool a cabinet unless air baffles are utilized to isolate the cold air and hot air sections. Without baffles, roof-mounted fan will draw not only the desired hot air in the rear, but also a volume of cold air from the front prior to being drawn in by the IT load. This serves only to cool the volume of hot air which we have previously established as a bad strategy. Similarly, providing a cut-out in the access floor directly beneath the cabinet will provide cold air to the inlet of the IT loads, however, it will also leak air into the hot aisle. Again, this only serves to cool the hot air.

Isolating high-density equipment

While isolating high-density equipment isn’t always a bad idea, special considerations must be made. Isolating the hot air is in fact, a good idea. However, the problem is in achieving a sufficient volume of cold air from the raised floor. Even then, assuming enough perforated floor tiles are dedicated to provide a sufficient air volume, too much of the hot air re-circulates from the back of the equipment to the front air inlet and combines with the cold air.

For more information on data center cooling, please download my newest White Paper, Data Center Cooling Best Practices, at http://www.ptsdcs.com/white_papers.asp. You can also view additional publications such as the following at our Vendor White Papers page:

Thursday, March 22, 2007

The New Data Center Boom

Across the country, data center development is booming. Companies, including major players like Microsoft and Google, are buying up acres of land with the intent of building new data centers.

This rapid growth is, at least in part, spurred by the requirements of the Sarbanes-Oxley Act and the Health Insurance Portability & Accountability Act (HIPAA), which call for better handling and storage of data. Companies are also responding to the nationwide push to establish energy efficient data centers. In order to accommodate the state-of-the-art, next generation data centers, companies simply need more space than their current facilities can provide.


Data Center Site Selection

For companies seeking to develop a new data center facility, high-quality site selection is of the utmost importance. By choosing a site location wisely, companies can save both time and money, while achieving scalability, flexibility and high availability.

Choosing a site that minimizes the natural and man made threats to continuous operation is the first step in provisioning a new data center. There are many factors to consider, including:

  • Natural Hazard Threats
  • Physical Location Threats
  • Terrorist Activity Threats
  • Environmental Contamination Threats
  • Site Accessibility
  • Amenities Access

It is interesting to note that the priority level of these factors is highly changeable. For instance, a decade ago it would have been more common for companies to seek site locations that with close proximity to major cities and airports. However, in the wake of September 11th, data centers are more likely to spring up in smaller cities, reducing the likelihood of damage from terrorist attacks, but most especially in those areas of the country that have the lowest operation costs including utility rates, land acquisition costs, labor rates, tax rates, and cost-of-living expenses.

To help navigate the complex process of site selection, many companies employ data center consultants for assistance in selecting an appropriate geography on which to locate their data center. Site selection services are the optimal way to ensure your mission critical facility is set up in both a location and a building that can support constant availability.

Thursday, February 22, 2007

Reducing Data Center Power Consumption

When it comes to data center power, less is clearly more. By reducing the amount of energy their data centers consume, companies can take a burden off electricity suppliers, protect the environment and increase their profits.

Many in the data center industry have already seen the light when it comes to reducing power usage. Technology companies are developing more efficient hardware, researchers are re-evaluating the possibility of converting to DC-power, electricity companies are offering financial incentives for data centers that significantly reduce their energy use, and corporations are revamping their data centers for maximum power efficiency.

This past December, Congress lent further support to the movement to reduce data center power consumption when it passed H.R. 5646 into law. The legislation calls upon the Environmental Protection Agency (EPA) to analyze the consumption of data center power by the federal government and private enterprises.

According to a report by Eric Bangeman of Ars Technica:

“The EPA’s study will fall under the auspices of its Energy Star program, which promotes the use of energy-efficient products. As part of the investigation, it will also consider incentives to encourage the deployment of more energy-efficient hardware in data centers.” The new legislation will help to raise awareness of data center power consumption and will spur the development of additional energy-saving solutions.

The government’s support of energy-efficient data centers creates a winning situation for everyone involved. The increased availability of Energy Star-rated technology, introduction of government-backed incentive programs, and growing public support for energy conservation make the decision to switch to energy-efficient technology an easy one.

Tuesday, February 20, 2007

Computer Room Design Tips

Computer rooms are an important component of the overall data center environment. Their purpose is to shelter network and server infrastructure as well as their related cabling, otherwise known as the computer room’s critical load.

In creating a secure and efficient computer room design, special consideration must be given to good planning and the implementation of the right technologies. The success of your design is dependent on the long-term scalability, flexibility and availability of your facility. Here are some computer room design tips to help your business optimize network performance, achieve its long-term availability goals and avoid costly problems in your computer room:

Power
In any mission critical environment, it’s important to provide adequate, scalable power for the load. Comprehensive load studies can produce a reasonable estimate of your facility’s power requirements. Once you’ve assessed the power needs of your computer room, conceptual and detailed planning can go forward.

Cooling
To design a computer room cooling system that operates effectively, you need a firm understanding of the amount of heat produced by the equipment contained in the enclosed space, along with the heat produced by other heat sources, such as conduction from adjacent spaces. Be sure to account for factors such as ceiling height, access floor depth, equipment layout and overall heat load.

Scalability
The design and construction of your computer room should meet the current technological needs of your business, while allowing for expansion along with the changing technology and business landscape. The use of modular systems, where the characteristics of the modules are known and the steps to add more modules are simple, is an excellent strategy to address growth without major disruptions.

Redundancy
High-availability is accomplished by providing redundancy for all, major and minor, systems, thereby eliminating single points of failure. By installing additional resources for system redundancy, hardware upgrades can be handled without fear of network failures. Incorporate redundant systems into your initial computer room design and continue to do so as your facility expands or upgrades its technology.

Monitoring
After your computer room is complete, the job of monitoring the IT and support infrastructure begins. Computer room monitoring is the vital last line of defense in achieving a high availability environment. When evaluating monitoring systems, look for solutions that are cost effective, easy-to-use, designed with intuitive alarming and escalation methodologies, and built to provide robust reporting all from a central, secure, locations.

Friday, January 26, 2007

Server Room Security Measures

The other day I was reading a news story about hollow coins being used for espionage and it inadvertently got me thinking about server room security issues. While I’m still not 100% sure of the best way to protect your facility against Canadian spy coins, I am aware of a number of techniques for guarding against unauthorized server room access.

To reduce downtime from accidents or sabotage due to the presence of unnecessary or malicious people, it’s important to implement server room security measures that account for a wide variety of potential threats. Whether building a new facility or renovating an old one, you’ll want to begin by mapping out your server room and identifying its most vulnerable areas. These may include access points, sensitive IT equipment and critical elements of the physical infrastructure.

Controlling Access to the Server Room

Server room security begins with controlling access to your facility. Security cards, biometrics and other auditable methods are commonly used to limit who is able to gain entry into the server room, but these methods can only do so much. Security cards, keys or passwords can fall into the wrong hands, while biometrics devices are expensive and may accidentally keep out people who should have access.

If these were your only options, it would be a tradeoff between lower security with convenience and higher security with hassles. By pairing either of these methods with backups such as IP-based camera surveillance, security guards or dry contact sensors, your server room is much better protected. Rather than relying on one strategy, a combination of security measures will provide the best result, particularly if they grow more stringent as you move toward the heart of the facility. By combining methods, you increase reliability.

Reinforcing Physical Infrastructure

From the ground up, the physical infrastructure of your facility should also contribute to your server room security. It pays to incorporate architectural and construction features that discourage or thwart intrusion. For example, make sure the walls of your server room extend past the ceiling, to the roof, to eliminate potential break-in points.

Reinforcing the physical infrastructure of your facility does more than just protect mission-critical IT equipment from theft or sabotage; it also gives protection to HVAC systems, power generators and fire suppression systems – anything that, if compromised, could result in downtime.

Securing IT Equipment

In addition to network security measures, it is important to implement physical security for IT equipment. Within the server room, rack-level security is a top concern. Rack locks defend against unauthorized access to critical equipment by limiting who can touch what. Not only does this help prevent sabotage, it also reduces the number of accidents and mistakes caused by workers interacting with technology that they may not be qualified to use.

Choosing a Security Solution

Every facility has its own unique security needs. When designing a security plan for your server room, carefully weigh your options. The goal is to find an acceptable compromise between security and its expense. By combining an assessment of risk tolerance with an analysis of available technologies and access requirements, it is possible to find an affordable, effective solution that will be accepted by users.

Thursday, January 18, 2007

Data Center Power Solutions

I receive a lot of questions from clients seeking solutions to their problems with data center power consumption. It seems that the higher energy costs rise, the more power the average data center needs. Overall costs for data center power may be skyrocketing, but there are ways to mitigate the expense. Here are some interesting suggestions I’ve come across lately:

- Utilize virtualization software
Virtualization technology is being trumpeted by many as a great way to get more bang for your data center buck. By means of virtualization you can reduce the number of servers that are required to run your applications, thereby increasing the operational efficiency of your data center.

This has become such a hot option that many manufacturers, including Intel, are now building virtualization capabilities into their chips. Companies such as Pacific Gas & Electric Co., which provides electricity to northern California, are jumping on virtualization as an opportunity to cut energy usage by offering financial kickbacks to data centers that save power after implementing virtualization technology.

- Switch from AC-power to DC-power
The idea of running data centers on DC-power isn’t a new one. Since DC requires fewer conversions than AC, there’s great potential for energy savings – some researchers predict a 10 to 20 percent reduction in power costs. Making the switch seems like a no-brainer, right? Unfortunately, higher engineering and technology costs keep DC-power from really catching on. It’s estimated that the cost for DC compatible equipment can climb up to 40 percent higher than that of AC-based technology.

- Install multi-core processors
Multi-core processors can give your data center’s hardware, as well as its energy savings, a boost. Multi-core processors may use slightly more power than a standard processor, but they run faster. This means one multi-core can do the job of several individual processors, reducing the amount of equipment and energy needed to get the job done.

- Optimize your cooling systems
Data center power and cooling go hand-in-hand. To optimize your CRAC systems and reduce energy consumption, focus on adjusting your air flow to eliminate hotspots. A/C units run most efficiently when operating at approximately 80% capacity and when they’re fed the hottest air. If you introduce additional cooling equipment without first trying to improve the efficiency of your existing setup, you’re doing your data center a disservice.

- Have a “Meeting of the Minds” between IT and Facilities Management
People in data center facilities management often complain that the IT team doesn’t properly consult them before purchasing new equipment, which leads to issues with data center power and cooling efficiency. The facilities management team has an intimate knowledge of the data center’s power and cooling infrastructure. By tapping into the combined experience of both departments, companies can sidestep potential energy wasters and keep the data center running at optimal efficiency.

For more solutions regarding data center power or other topics, please visit our Vendor White Paper archive.