Home | Internet | Internet Tools & Resources

Dollars, Cents, and Gigawatts: Online Tools Bind Great Power and Great Responsibility

By: Justin Blumling

Read More About Internet Tools & Resources

42's Data Center Calculators illustrate the dramatic cost and carbon savings through efficiency efforts.

IBM has built an extensive marketing campaign around "on-demand." Though the slogan has been more directed at their business patrons, the concept has become extremely relevant to our personal lives. We can have an entire musical catalog in minutes. Tracing our family histories doesn't require exhaustive research, only points and clicks. Forget the mall on Black Friday; Amazon, Apple, and Abercrombie.com are always open. A generation of young scholars need never experience card catalogs, microfiche, or the Dewey Decimal System.

These web services-music downloads, shopping, research-were once a novelty. Now they are globally pervasive and presumed to be as automatic as death and taxes.

But there are consequences to this on-demand lifestyle. Few think about the bevy of servers to process this data or the vast storage systems to house it. Fewer think about the energy needed to run this equipment perpetually-24x7x365. Even fewer appreciate the costs for all parties involved, including our ecosystem. Research suggests a single server operating 24 hours a day, 7 days a week, 365 days a year, emits as much CO2 as a Toyota Camry driving 15,000 miles per year. There are thousands of servers in thousands of server rooms, labs, and data centers across the country.

The Gigawatt Age In 2006, peak demand from US data centers was estimated at seven (7) gigawatts. One (1) gigawatt represents one million kilowatts (kW) or one billion watts (W).

This is but one of the remarkable statistics concerning servers, data centers, and energy:
The average Data Center consumes as much energy as 25,000 households (Parrino, 2009) * Data Centers are expected to surpass the Airline Industry in greenhouse emissions by 2020 (Parrino, 2009) * In Northern California, the home of Silicon Valley, data center energy consumption amounts to 400 megawatts to 500 megawatts (Scofield & Weaver, 2008). The consumption is so concentrated that Pacific Gas and Electric, a utility local to Silicon Valley, has presented six-figure rebates for companies that design for efficiency day one. * The majority of data centers are overcooled and run far less efficiently than they could be.

The Environmental Protection Agency, in a 2007 report to Congress, relayed the most staggering numbers. Data center energy use in 2006 hit 61 billion kilowatt hours at a cost of $4.5 billion. By 2011, these numbers could hit 100 billion kilowatt hours and $7.5 billion if the industry takes a laissez-faire approach to efficiency. Though these servers are mostly out of sight, they are certainly not out of mind for industry authorities. Attend any data center conference or tradeshow and you are guaranteed to find "efficiency" and "green data center" topics in the program.

The attention is warranted. The continued growth of computing in our professional and personal lives begs the question: If consumption is extreme now, what may lay ahead in the future?

The Business Case Despite this sobering news, a simple fact remains: businesses are businesses. While environmental responsibility and stewardship are great for press releases and corporate web sites, most businesses require a cogent business case to undertake efficiency improvements, especially in light of a poor economy. Coupled with the traditional disconnect between IT/Facilities and a lot of new terms and metrics, many data center professionals may not know how and where to begin efficiency efforts.

To help data center professionals tackle the important but complex topic of efficiency, 42U has developed two, free interactive calculators, available on 42U.com. These online tools hope to align all stakeholders in an ongoing data center efficiency program, where significant savings and operational improvements are possible.

PUE/DCiE Calculator Power Usage Effectiveness (PUE) is a new term in the data center lexicon, comparing a data center's total energy requirement to the energy consumed by the IT load. A smaller result is better, as it means more energy is devoted to the IT process (the denominator) while less energy is lost in conversion and support loads (minimizing the numerator).

For example: Total Facility Load: 200kW/ Total IT Load: 120kW = 1.66 PUE Data Center Infrastructure Efficiency (DCiE) is simply the inverse of PUE, allowing efficiency to be expressed as a percentage: Total IT load: 120kW/ Total Facility Load: 200kW x 100 = 60% DCiE.

With 42U's PUE Calculator, a data center operator does not need to know the origin or intricacies of PUE. If he can access his IT load (from the UPS), and his facility load (from the meter), he receives an initial PUE score.

The calculations then go a step further. Using the PUE score, the built-in energy function estimates total annual energy use, annual energy expense, and annual carbon footprint. For data center professionals who don't have energy bills or reports readily available, the calculator provides a brief snapshot into the energy profile of their facilities.
Energy Savings Calculator For more detailed calculations, 42U's Energy Savings Calculator makes use of the same formulas to project energy and CO2 savings over one, five, and ten years.

Users start with their current PUE and then set a goal PUE- a lower value that's the result of improvements relevant to their facility. The difference between current PUE and goal PUE accounts for the potential energy and cost savings.

Simple math shows that modest improvements can have considerable effect. A 1 megawatt facility (total load) at $.08 per kWh can save over $70K in energy costs and 528 Tons of CO2 emissions in going from a 2.0 PUE to a 1.8 PUE.

If data center professionals are contemplating efficiency improvements, they can compare the cost of proposed solutions with their energy savings to determine their payback period.

Equally if not more important are the potential capital expenditure savings (CAPEX). A more efficient data center reduces the need for additional equipment assets and potentially eliminates the need for construction projects. The CAPEX savings, in avoiding construction, have proven to be 5-6 times the OPEX savings achieved in making a data center more efficient, not to mention the underlying CO2 or GHG savings.
Conclusion PUE is a key ingredient in these online tools. Though not everyone has calculated it, the metric has perhaps gained the most traction of any in the industry. One of the reasons it succeeds is simplicity.

With data centers, one-size-fits-all is uncommon. But every data center requires power to carry out its function. Every data center, by definition, has a portion of that power dedicated to IT equipment. PUE simply compares these values.

While PUE appears to be universal as a formula, the results are a different story. What one data center can achieve may not be relevant to another. For instance, PUE is a dynamic value that varies over time and with workloads. A facility in its infancy, with nominal loads, would be hard pressed to obtain an optimal PUE. Then there's Google, who continues to rewrite the book on data center design and operation. They recently unveiled a data center in Belgium that operates without chillers. On those days when it's too hot outside, Google can shift that data center's workload to another-a distinct advantage of a comprehensive, mirrored data center network. Not every enterprise has that luxury.

So, while the cloud giants of the world publish and promote PUE numbers that hover between 1.15 and 1.25, we encourage data center professionals to take a critical look at what they can do to optimize their unique space. No one shares the exact same layout, equipment, applications, floor space, and set of challenges. Figure out what efficiency program aligns with your facility, your budget, and your goals. These online tools provide an excellent starting point for those conversations.

Article Source: http://depositarticles.com/

Justin Blumling brings five years of data center experience to his role as lead Product Support Manager for 42U.com sales, consulting and marketing team. In addition to his technical expertise, he writes original content including PUE Calculator and Energy Savings Calculator solutions and collateral pertinent to data center trends and technologies.

Please Rate this Article

 

Not yet Rated

Click the XML Icon Above to Receive Internet Tools & Resources Articles Via RSS!
Additional Articles From - Home | Internet | Internet Tools & Resources
Sign Up for a free account.
Our other websites
All about SEO
Beauty Tips Forum
Submit Articles
Member Login
Link Us
Submission Guidelines
Ezine Notifications
Article RSS Feeds
Sitemap
New Stuff
About Us
Developer
Good shopping portal
Link to Us
Contact Us
Privacy Policy
Terms of Service

counter easy hit

Powered by Article Dashboard