- Harnessing datacenter heat for savings
- Xerox develops Sustainability Calculator for doc tech
- Make IT accountable for tech-related power bills
- Ecobutton a quick, slick reminder to power down PCs
- Report: Green tech could save Feds nearly $275M per year
- Dell's shiny green blade server
- Emerson delivers free Energy Logic blueprint for building a power-efficient datacenter
- Q&A: From 90 years old to LEED gold
- Govs pledge to buy greener hardware for state governments
- IBM aims at trimming electric bills with Active Energy Manager
April 03, 2008 | Comments: (0)
Harnessing datacenter heat for savings
The heat that pours out of your datacenter machinery represents a hefty chunk of your monthly utility bills. After all, were it not for that heat waste, you wouldn't be pouring dollars into running that pricey cooling equipment to ensure your valuable hardware doesn't get fried.
While many vendors out there are devising CRAC hardware and cooling technologies for more efficiently and inexpensively beating the datacenter heat, some organizations are taking another tack: putting all of that hot air to valuable reuse, which can have pleasant financial and environmental benefits.
Datacenters are responsible for a goodly amount of heat waste, to be sure: "In many cases, a datacenter can generate enough heat to heat a building 10 to 30 times its size," says says Steve Sams, vice president of IBM Global Site and Facilities Services. "That's a lot."
Recycling heat
One of the most logical uses of datacenter heat waste is, indeed, to warm up the rest of an organization's building. Intel is doing just that with a heat-reuse system designed for its first LEED (Leadership in Energy and Environmental Design)-certified green building, a development center in Israel.
The system employs HR (heat recovery) chillers instead of standard chilled water systems. While a chilled water system releases hot condenser water directly into the atmosphere via the cooling tower, the HR chiller system uses that heated water to warm the rest of the building in the winter and provide hot water for bathroom and kitchen use year-round.
According to Intel, "reusing datacenter heat eliminates the need to add boilers for heating the rest of the building. This energy-saving approach accumulates points towards environmental certification and is highly cost-effective."
The company projects savings of approximately $235,000 annually due to reduced fuel consumption, resulting in an ROI of just over 19 months.
Sharing the warmth
Of course, not every datacenter has a nearby facility to heat up -- but that doesn't mean the excess heat it produces needs to go to waste. Co-location service provider GIB recently partnered with IBM to transform a former underground military bunker in Uitikon, outside of Zurich, into a highly secure data-storage facility. Operating at full capacity, the facility is expected to create 2,800 megawatts of wasted heat per year (the same amount of energy needed to supply up to 80 houses with heating and warm water for one year).
"Whenever we can, we encourage our clients to find opportunities to reuse waste heat in the datacenter environment," says IBM's Sams. In this case, however, there was no office building to reuse the facility's heat, given that it's out in the middle of the woods and hidden underground.
Turns out there was a better alternative to simply blowing cold air at the problem: GIB and IBM devised a direct heat exchange between the datacenter and local public swimming pool. Yes, the people of the town can now enjoy a swim in water warmed up for free by the datacenter's heat waste.
The green payoff here is pretty evident. The town is saving money by longer paying to heat its pool. GIB earns CSR (corporate social responsibility) points. As for the environmental benefits: "Through reclaiming the heat, approximately 130 tons of carbon emissions can be saved. This corresponds to the CO2 discharge of mid-size cars driving 500,000 miles," Sams says. "It's a nice solution. It's obviously a terrific example of the private sector and the public sector working toward each other's mutual benefit."
In addition, the approach demonstrates that you don't need to start from scratch with a new datacenter facility or office building to make use of heat waste. The public pool clearly had its water-heating system in place before GIB and IBM built this new datacenter -- yet they were still able to connect the two to reap the aforementioned benefits.
Talkin' 'bout co-generation
Here's one more example of an organization harnessing its heat waste for reuse: NetApp, based in Sunnyvale, Calif., has a rather different approach to reusing heat waste: cooling its datacenter hardware.
In NetApp's case, that heat isn't coming from its IT equipment, it's coming from its natural gas-powered co-generation system. The system is used to produce electricity for the 1MW datacenter -- though only when the temperature and energy prices in California peak. When the system is running, it produces heat waste, which NetApp uses to power its adsorption chiller that chills the water used in its cooling system.
According to NetApp, the co-generation system has an overall efficiency rating of 75 percent to 85 percent (compared to the 35 percent efficiency for a natural-gas utility). NetApp says its co-generation system saves the company $300,000 annually.
We'll no doubt continue to see approaches to reusing heat waste evolve. Perhaps, soon, there will be a way to efficiently transform it into power for the datacenter. Even today, though, there's already a potential ROI, both in terms of economic and environmental benefits.
Posted by Ted Samson on April 3, 2008 03:00 AM
March 24, 2008 | Comments: (0)
Xerox develops Sustainability Calculator for doc tech
Add Xerox to the list of vendors adding green-o-meter functionality to their wares. The company Tuesday will unveil what it dubs a Sustainability Calculator, designed to help customers evaluate the environmental impact of their document-technology systems, such as printers, faxes, and copiers.
Part of the company's Xerox Office Services, the Sustainability Calculator measures the waste and greenhouse gas emissions associated with powering printers, copiers, fax machines, and multifunction devices. It also measures the differences, in environmental terms, resulting from practices such as printing single-sided documents instead of double-sided, or using different types of ink.
In addition to the aforementioned version, which Xerox reps would use during an assessment for a customer, the company also has developed a slimmed-down Web-based version.
Both calculators require a user to input information about the various machines in his or her organization. They then employ "proprietary algorithms and document assessment research to deliver data about a company's entire fleet of office products, from printers to multifunction devices and copiers, regardless of the equipment supplier," according to Xerox.
Once a customer has a glimpse of the inefficiency of its document-tech systems, the next step (Xerox hopes) is to show customers how to gain efficiency by, say, in retiring various older copiers, fax machines, and printers for fewer, newer MFDs.
It might be tempting for to dismiss this type of tool as simply a marketing scheme to exploit CXOs who've caught a case of the green fever that's swept the U.S. and beyond. But the reality is, there are cost savings to be had from certain sustainable practices, and making adjustments to your company's network of printers and other document-technology products is one of them. That includes moving to fewer multi-purpose machines as your older ones are ready for retirement.
First, a new model MFD should be Energy Star compliant, which means it has a significantly lower power draw -- as much as 70 percent, according to Xerox -- than that of its one-function predecessors combined. Also, from a green perspective, manufacturing and shipping four machines -- a copier, a scanner, a fax machines, and a printer -- requires more resources than does building and shipping a single MFD that can do the work of four.
According to Patricia Calkins, vice president of environment, health, and safety at Xerox, many customers have been asking for more information about the green benefits of consolidating doc-tech systems and improv-ing their printing practices. "I was very surprised as I've been doing customer roundtables," says Calkins. "We talk about optimizing the office, and people have said, 'We understand the financial benefits. We want to talk about the environmental benefits."
Global defense and technology company Northrop Grumman worked with Xerox at one of its sectors to reduce a fleet of 2,000 printers, hundreds of MFDs, and stand-alone copiers to fewer than 1,100 devices. According to the Sustainability Calculator numbers, the change resulted in a savings of 27 percent in energy usage while reducing GHG emissions by 26 percent and solid waste creation by 33 percent.
Posted by Ted Samson on March 24, 2008 09:55 PM
March 13, 2008 | Comments: (0)
Make IT accountable for tech-related power bills
If more IT managers were directly responsible for their department's energy bills, they'd be more likely to implement greener, energy-saving practices.
That's one of the take-aways from a recent survey commissioned by 1E, developer of a PC power-management product called NightWatchman. Among 100 IT managers at enterprise organizations in the U.K., 90 percent said they were aware that their employers have broad environmental policies. However, around one-third of the survey respondent said they felt zero pressure to reduce power consumption.
1E attributes this seeming discrepency to the fact "that only 23 percent of IT managers surveyed by 1E have direct ownership of their corporate power bills."
"1E has always believed that only by giving IT teams direct ownership of corporate power bills will they feel truly empowered to address this globally important issue. The findings of our latest research survey support this view," said Simon Francis, 1E's VP for Energy Solutions, in a written statement.
I'd say there's indeed something to 1E's assessment of the data. Many vendors and customers I've spoken to in regard to sustainable-technology practices have noted that there's a disconnect between IT departments and whomever at their respective companies is responsible for paying the energy bills, e.g. the facilities department.
As a result, IT admins tend to have have what amounts to a blank check when it comes to powering an organization's IT equipment, from the desktop systems to the datacenter hardware. But what would happen if the IT manager was suddenly held accountable for those costs and was given incentives to reduce energy consumption from month to month? The resourceful IT manager would no doubt find ways to do so. Heck, even the not-so-resourceful one would find ways; there's plenty of them, from the desktop to the datacenter.
For more on the subject, check out this a fine piece by James Murray over at BusinessGreen blog.
Does it makes sense for your IT department to be responsible for energy bills associated with IT operations? Let me know what you think.
Posted by Ted Samson on March 13, 2008 04:07 PM
January 28, 2008 | Comments: (0)
Ecobutton a quick, slick reminder to power down PCs
Not only are tradeshows growing greener; so to is the promotional swag companies are doling out. One fine example is a clever device dubbed the ecobutton from called a company called BIG.
The concept behind the ecobutton is pretty straightforward. You connect the device to your computer via USB and download the necessary software. Then the big green illuminated ecobutton sits atop your workspace, reminding you how much you like nature and/or dislike wasting energy (or paying to waste it). Whenever the time comes to take a break -- short or long -- you simply press the button. Doing so puts your system into what BIG deems ecomode, "which ensures that both your computer and monitor draw only the same nominal power as when they are shut down."
When you return from the break, you press the ecobutton again and voila, your system awakens instantly, according to BIG. (I'd love to actually see for myself whether it can rouse my machine instantly, as opposed to "eventually.")
As a bonus, the accompanying software lets users see how much money, power, and carbon units they've saved using the device. PC power management tools like this can save you as much as $50 per year in energy bills. (By the way, the ecobutton is also ROHS compliant.)
What makes the ecobutton promotional tool? Companies can order ecobuttons in bulk and have their company logo, message and Web address incorporated into the ecobutton splash screen.
It's a neat idea, this ecobutton. Some might argue that it's a bit too ironic to promote waste-reduction by attaching yet another piece of plastic and metal to your machine, and that there are tools available that enable you to put a machine to sleep with a couple of mouse-clicks. However, I see a positive trade-off here: The brightly illuminated green button should serve as a constant visual reminder to users to not only to power down their systems but to hopefully conserve in other ways.
The ecobutton starts at around £12.76 ($25), and bulk discounts are available. Go to the ecobutton Web site for more information.
(Thanks to Environmental Leader for the find.)
Posted by Ted Samson on January 28, 2008 10:44 AM
January 24, 2008 | Comments: (0)
Report: Green tech could save Feds nearly $275M per year
Efficiency doesn't tend to be a hallmark of the federal government, which might explain why the words government contract induces salivating amongst some CXOs the way bells caused Pavlov's pups to slobber.
Green technology stands to reduce some of that wasteful tax-fueled inefficiency, a proposition presented by HP and Intel in a couple of reports titled "Go Green Power Play" and "Go Green PC Power Play" aimed at the U.S. federal government.
The short of it: Were the Feds to upgrade all of their PCs, monitors, and servers to energy-efficient alternatives, and were they to adopt some power-saving technologies like virtualization, they could save taxpayers around $275 million in energy costs per year, according to the reports from HP and Intel. In the process, the sustainability move would ease the pressure on the nation's overstrained power grid by one billion kWh per year.
Breaking it down, HP and Intel assert in "Go Green Power Play" that the Feds could drive down their datacenter energy costs, estimated at around $479.5 million annually, by as much as 40 percent (that is, around $192 million) by employing various green technologies. Those include virtualization, consolidation, better cooling technology (specifically HP's own Dynamic Smart Cooling and Thermal Mapping technologies), more energy-efficient IT systems (specifically Intel-powered machines from HP), and power-distribution efficiencies.
The figures are based on IDC's estimation that the current installed server base of the U.S. Department of Defense and civilian agencies is just under 600,000, that each server currently consumes 2,000 kWh per year, and that the average energy cost is 10 cents per kWh.
In the PC-oriented "Go Green" report, HP and Intel suggest that the Feds could save more than $82.4 million annually in energy costs by upgrading all their Energy Start 3.0-compliant desktops, laptops, and monitors to Energy Star 4.0 models. Activating power-management features to ensure machines power down when not in use plays into the savings equation.
Here, the report assumes Feds are powering around 4.5 million desktops, 440,000 CRT monitors, nearly 4 million LCD monitors, and just over a million laptops.
A final point in all this: The Feds aren't the only ones who could benefit from investing in some of those green-tech upgrades. This kind of energy waste is rampant at organizations large and small. It certainly scales impressively in the context of the government's vast fleet of systems, but organizations of all sizes could stand to assess their respective situations. There's plenty of potential gain, both monetary and environmental, to be reaped.
Posted by Ted Samson on January 24, 2008 04:40 PM
January 24, 2008 | Comments: (0)
Dell's shiny green blade server
The PowerEdge M-series from Dell has it all, from a hyper-efficient power supply to superior cooling and power management
From a green-tech perspective, the most intriguing server I've ever known (from a distance) was the Gemini Green Series from Open Source Solutions (OSS). Highly efficient power supplies? Check. Efficient internal cooling? Check. Easily swappable components -- from motherboards to memory to power supplies -- to eliminate the need for ripping and replacing upgrades? Check.
Alas, OSS is no more, but its legacy won't be forgotten -- not by me, anyway. The fact that it's gone the way of rainbow suspenders is no reason to dismiss the beauty of its product design either. In fact, I see glimmers of the Gemini in Dell's newly announced PowerEdge M-Series chassis and blades. Dell designed the M-Series "from the ground up using Dell Energy Smart technologies, resulting in 30 industry patents," clearly with sustainability in mind.
Playing it cool
Like the OSS Gemini (which was a 2U server, by the way), Dell's new PowerEdge blades employs 90-plus percent efficient power supplies, developed in-house. (For reasons I still can't fathom, the standard among server vendors still appears to be in the realm of 80 percent.)
Not only is Dell's power supply capable of hitting a higher energy-efficiency level; it does so more quickly than rival power supplies, according to Mike Roberts, senior product planning manager for the M-Series line. He says that most power supplies achieve their maximum level of energy efficiency only when the supply is running at 90 to 100 percent utilization -- which certainly isn't the norm. "We get to a really good efficiency level at relatively low threshold, 88 percent at only 20 percent utilization," he says.
Those seemingly small differences among power supplies can actually make a dramatic difference in terms of reducing a server's power and heat waste -- and the associated costs scale impressively the larger your datacenter.
Dell also claims to have developed an innovative internal cooling system, designed to adapt to the needs of both high-end and low-end configurations. In addition to its optimized fans (as well as their underlying algorithms), the system boasts a superior airflow design, according to Roberts; Dell has taken great pains to remove impedance throughout the chassis. "The easier the air flows, the less hard the fans have to work," says Roberts.
The chassis also has three distinct cooling zones, each cooled by its own fan bank. In lower-end configurations where the chassis isn't fully loaded with blades, "the fans on the side can run really slowly because they don't have to work hard to cool their zone," says Roberts.
There's also the swappability factor (a word you will not find in Webster's, by the way). In an ideal green-tech world (mine anyway), an IT admin would be able to swap in and out all major components in his or her servers, while they're on the rack, such that machines wouldn't need to head to the shop or, more likely, to the recycling bin if a substantial upgrade is needed. It doesn't look like the major hardware vendors of the world are ready to offer that level of hardware interoperability, but here, Dell has taken a step in that direction.
Specifically, the company has developed what it dubs FlexIO switch technology for easily upgrading the machine's network connectivity up to 10Gig without replacing the base switch. For the enclosure, customers can opt for an upgradeable Dell PowerConnect M6220 Layer 2/3 Ethernet blade switch, with 1Gb ports and optional bays that can support either 10GbE or stacking ports.
Further, customers have three Cisco Ethernet switch choices, including a switch with a variety of 1Gbps, 10 Gbps, and stackable ports. Also available: a Cisco Infiniband switch. Add to that the options for two Brocade 4Gbps Fibre Channels, as well as Fibre Channel and Ethernet Pass Through options.
Additionally, with an eye on the future, Dell has designed the M1000e enclosure not to be dependent on specific server processor/chip set architecture. Further, it will be able to accommodate double-wide blades down the road.
Secret software sauce
Dell's hardware choices alone don't result in a more energy-efficient machine. Like an increasing number of vendors, Dell is turning to server management software to rein in energy waste -- a promising development that stands to lower those power bills and extend the life of hardware.
In Dell's case, that software takes the form of its Version 5.3 of its OpenManage systems management suite, released last November. Dell touts the package -- which comes at no cost with the blade package -- as "easy-to-use yet powerful management tools that help reduce the cost and complexity of managing computing resources."
Among its features is dynamic power management, which enables admins to set high- and low-power thresholds to help ensure blades operate within their defined power envelope. The norm is for servers to consume the maximum recommended amount all the time, even if they're not usually being run at full bore. Moreover, the package offers real-time reporting for enclosure and blade power consumption, and the ability to prioritize blade slots for power to provide optimal control over power resources.
So, for example, if an admin were to allot 3,000 watts to a given chassis, the system would distribute power evenly among all the blades. However, if the chassis wasn't pulling enough power, for whatever reason, it could be set to prioritize which blades would be throttled down first.
Your benchmark or mine?
As Dell tells it, the various greenovations it's injected in these babies reap superior power efficiency (that is, performance per watt) over rival blade offerings from HP and IBM. According to a Dell-sponsored study by Principled Technologies, "the PowerEdge M-Series consumes up to 19 percent less power and achieves up to 25 percent better performance per watt than the HP BladeSystem c-Class. Compared to the IBM BladeCenter H, the M-Series consumes 12 percent less energy and achieves up to 28 percent better performance per watt."
Were you to apply a cost-per-kilowatt amount to those figures, as Dell has, you'd save $2,600 annually per year over HP's blade competitor and $1,500 per rack per year over IBM's.
I wouldn't discount Principled Technologies figures outright, just because the study was sponsored by Dell. It's entirely conceivable that, given the power-efficient measures Dell has taken, its blades consistently delivered better per-watt performance -- but only in that test, which used the SPECjbb2005 benchmark.
Problem is, that benchmark wasn't really developed to measure power efficiency. Organizations and analysts are still grappling as to what method is best for accomplishing that, as evidenced by a recent study comparing the power efficiency of AMD and Intel's respective quad-core chips.
The point is, benchmarks are slowly emerging, such as SPEC's recently announced SPECpower_ssj2008. However, that benchmark isn't suited for a blade environment. In short, work here remains to be done.
Hopefully the InfoWorld Test Center will have a chance to test Dell's claims against offerings from HP and IBM. Whatever the outcome of that test might be (again, if it happens), I still extend kudos to Dell for further raising the green bar more among hardware vendors.
One last thought: IBM and HP may claim that they're not worried about Dell's latest foray into the world of blades. In fact, a rep from one of those vendors sent me a rather dismissive note prior to the official Dell announcement: "We've been hearing rumors over here that Dell finally plans to announce their copycat blade on Monday. Odd that they'd choose a national holiday to announce, but that's another story ;-)."
Of course, when you go out of your way to eagerly point something out and declare you're not worried about it -- you're probably at least a leetle bit worried about it. Conceivably, Dell has surprised the competition with the level of innovation it's brought to the table here, enough to generate some concern.
Posted by Ted Samson on January 24, 2008 03:00 AM
November 29, 2007 | Comments: (0)
Emerson delivers free Energy Logic blueprint for building a power-efficient datacenter
How-to guides are downright handy for tackling daunting projects, such as deploying a new BPM (business process management) solution or building a birdhouse (one of those really fancy ones with indoor plumbing).
One of the most daunting tasks that companies face today is figuring out how to wring greater energy efficiency out of their datacenters. Certainly, many vendors are ready to step up and demonstrate where their respective products fit into the power-saving puzzle. But there's something to be said for a vendor-neutral blueprint to plan the overall task.
That all is a wordy lead-in to pointing you to a new report -- available as a free download -- released today by Emerson Network Power. It's titled "Energy Logic: Reducing Data Center Energy Consumption by Creating Savings that Cascade Across Systems," and it's an impressive piece of work, outlining ten interrelated technology strategies that comprise a holistic approach to improving datacenter energy efficiency by as much as 50 percent, according to the company. And as I noted, it's free.
For a little background, Emerson has coined the term "Energy Logic" in this report, a strategy which, according to the company, "centers on 'the cascade effect' by which one watt saved at the processor level can save an average total of 2.84 watts in energy consumption."
The report starts at the server component level, outlining the benefits of low-power processors. "Independent research studies show these lower-power processors deliver the same performance as higher power models," the report says. "In the 5,000-square-foot datacenter modeled for this paper, low-power processors create a 10 percent reduction in overall datacenter power consumption."
Next up: power supplies. The report notes that most power supplies found in servers are working at around 72 percent effiency -- yet "best-in-class power supplies are available today that deliver efficiency of 90 percent. Use of these power supplies reduces power draw within the data center by 124 kW or 11 percent of the 1127 kW total," the report says.
From there, the report suggests that datacenter operators look at power-management software. Despite the fact that processors have built-in power-management features, they end up disabled for fear of crippling response time. Yet "in idle mode, most servers consume between 70 and 85 percent of full operational power."
The Emerson Energy Logic report suggests that admins reconsider how they use power-management features. In the 5,000 square foot datacenter model, the report says that using power-management features can reduce peak power draw from 80 percent to 45 percent, saving "an additional 86 kW or eight percent of the unoptimized datacenter load."
Blade servers have a role to play in the energy-efficient datacenter, according to the Emerson report. "Blade servers consume about 10 percent less power than equivalent rack-mount servers because multiple servers share common power supplies, cooling fans and other components. ... More importantly, blades facilitate the move to a high-density data center architecture, which can significantly reduce energy consumption."
Moving on, Emerson's Energy Logic strategy highlights server virtualization. In the 5,000 square foot model, "assuming 25 percent of servers are virtualized with eight non-virtualized physical servers being replaced by one virtualized physical server, ... virtualization provides an incremental eight percent reduction in total datacenter power," according to the report.
No. 6 on Emerson's energy-efficiency menu for datacenters: best cooling practices. That includes "sealing gaps in floors, using blanking panels in open spaces in racks, and avoiding mixing of hot and cold air." This is low-hanging fruit that requires no additional technology investment but can result in a five percent efficiency boost, based, again, on the 5,000 square foot datacenter model.
415V AC power distribution is the next strategy component in Emerson's Energy Logic scheme. The short of it is, most UPS systems are rather inefficient, as they convert incoming power to DC and then back to AC within the UPS. "In most datacenters, the UPS provides power at 480V, which is then stepped down via a transformer, with accompanying losses, to 208V in the power distribution system," according to the report. "These stepdown losses can be eliminated by converting UPS output power to 415V." The result: "an incremental two percent reduction in facility energy use" in the 5,000 square foot model.
Cooling reappears in the report next, specifically variable capacity cooling. "Typically, CRAC [computer room air conditioners] fans run at a constant speed and deliver a constant volume of air flow. Converting these fans to variable frequency drive fans allows fan speed and power draw to be reduced as load decreases," the report says. Emerson specifically cites digital scroll compressors here, which "allow the capacity of room air conditioners to be matched exactly to room conditions without turning compressors on and off."
The payoff: "A 20 percent reduction in fan speed provides almost 50 percent savings in fan-power consumption."
Ninth in the Emerson Energy Logic lineup is, lo, another cooling strategy: high-density supplemental cooling. Datacenter operators are cramming more machines into their facility, and CRAC systems alone can't handle the extra heat. "Supplemental cooling units are mounted above or alongside equipment racks, and pull hot air directly from the hot aisle and deliver cold air to the cold aisle," the report explains. These units can reduce cooling costs by 30 percent, Emerson reports.
Last but not least, there's monitoring and optimization. With varying types of hardware spread out around the datacenter, cooling can prove inefficient. "Cooling control systems can monitor conditions across the datacenter and coordinate the activities of multiple units to prevent conflicts and increase teamwork," the report says. "In the model, an incremental saving of one percent is achieved as a result of system-level monitoring and control."
There's plenty more information to be gleaned from Emerson's 21-page Energy Logic report, along with helpful charts and diagrams to help datacenter operators as they venture toward a greener, more sustainable facility.
Did I mention the report is available as a free download? Get it right here.
Ted Samson is a senior analyst at InfoWorld and author of the Sustainable IT blog, tracking trends toward greener, more energy-efficient IT. Subscribe to his free Green Tech newsletter here.
Posted by Ted Samson on November 29, 2007 10:02 AM
November 08, 2007 | Comments: (0)
Q&A: From 90 years old to LEED gold
There's a tendency in our society to simply throw out the old and replace it with something shiny and new. Sometimes, that might be necessary; often, though, the old still has value, and chucking it results in costly and unnecessary waste. An older server, for example, could still function as a test machine for patches. Or a 90-year-old building could be transformed into an energy-efficient, eco-friendly datacenter.
Skeptics may scoff at the notion, but not Jim Smith, vice president of engineering at Digital Realty Trust. Smith took the lead in transforming 20,000 square feet of a facility built in 1917 into the world's first LEED (Leadership in Energy and Environmental Design) Gold-certified datacenter.
Judging by previous interactions I've had with Smith, I can safely say that this project wasn't simply a "green-washing" publicity stunt. Smith approaches sustainable IT from a very practical, business-oriented standpoint: boosting efficiency and reducing waste results in long-term cost savings. Embracing LEED benchmarks, as companies such as Adobe, Qualcomm, and HP have done, helps to lay down the foundation for reaping those benefits.
Following is a Q&A between Smith and me about the datacenter renovation project.
Ted Samson: I see the building was built in 1917. Did the company recently acquire it and do a full renovation before starting to use it, as opposed to using it, then deciding to do the LEED renovation?
Jim Smith: We've owned the 350 East Cermak building in Chicago for more than two years and LEED Gold certified a new datacenter within the building as a project for a customer. We did not have to do any major renovations to get the LEED certifications, but it was nonetheless an intensive process. Not simple, but we did it and are proud of how it turned out.
The LEED Gold certification is not for the entire building. 350 East Cermak is a huge building -- more than a million square feet in total -- and this specific certification is for about 40,000 square feet, which is still a large space but not the entire building. We have another project in the works that will be an entire floor and we're excited about that.
TS: What was the thinking in upgrading an existing, old building as opposed to finding a more modern building or starting from scratch?
JS: A few factors went into the decision to use an existing building. One is that the customer really liked the 350 Cermak facility because it is such an ideal spot, right in the Loop and at the exchange. Another important factor for us is that the LEED certification encourages reuse of existing structures, which is very energy-efficient when you look at how much energy is used in materials and construction for new buildings.
There are some misconceptions that LEED certification only makes sense for new buildings, but this project showed us that the process definitely works for existing buildings. Even huge, old, historic structures like this one in Chicago.
One key takeaway from this project for the datacenter industry as a whole is that energy-efficiency initiatives should not just be limited to new construction. You can do great projects in existing buildings and get amazing results.
TS: Could you please give me a couple of "before and after" examples as to what changes you made to the facility to help it earn its Gold certification? I'm particularly interested in anything that improves energy efficiency, that was solved with IT tools, or that affects the datacenter performance.
JS: One important thing we used are sophisticated energy-measurement tools.... These are not expensive, but they provide critical data that let you understand what is happening in the datacenter. Every datacenter should have this, particularly since it is such a small investment and provides such valuable information. Step one for energy-efficient operations is always to have a way to measure.
Another important thing ... is commissioning -- both fundamental commissioning and advanced commissioning.... It was an important step in the process for this LEED Gold datacenter. The commissioning process is very meaningful because it verifies the design basis and makes you test the systems for efficiency. You learn a lot during this process and can apply that knowledge right upfront to make adjustments that ensure you are maximizing your energy savings.
We [also] installed tools to monitor outside air at the facility. The equipment and monitoring system makes sure the air is clean, and helps us improve performance of the ventilation system and improve indoor air quality.
TS: How much did the project cost?
JS: The financial impact of the LEED process was an extra 4 percent in cost -- half of which was materials and tools, and the other half of which was administrative. Very efficient from a cost perspective, and it has an overwhelmingly positive net present value. Definitely worth the cost.
One important thing that I would highlight is this: There is some skepticism out there about how relevant LEED certification is to datacenters and about how significant the benefits are. I would argue that is does have a lot of value and should be strongly considered as one of number of tools that companies can use to make their datacenters more efficient. It's not the end-all-be-all, but it does provide a way of focusing the engineering and design and construction and operations teams -- with a process that builds good habits and keeps them focused on objectives that do achieve energy efficiency.
Anyone who has ever been involved in a datacenter project knows the value in having a process that keeps those teams focused and organized in this way. That alone makes this process valuable, even before you factor in the energy-efficiency gains.
Posted by Ted Samson on November 8, 2007 03:00 AM
November 07, 2007 | Comments: (0)
Govs pledge to buy greener hardware for state governments
If you're a hardware vendor and "energy efficiency" isn't high on your product-feature list, you're risking losing some valuable customers -- perhaps as many as 50, with names such as Kansas and Minnesota.
The National Governors Association (NGA) today announced a partnership between the NGA chair's Securing a Clean Energy Future (SCEF) initiative and the Climate Savers Computing Initiative (CSCI) to spur deployment of more energy-efficient computers and servers in state offices and agencies.
Two governors, NGA Chair Gov. Tim Pawlenty of Minnesota and Kansas Gov. Kathleen Sebelius, have committed to the partnership thus far, and they will work to encourage their 48 peers to follow suit.
The move would certainly have eco-friendly benefits. Moreover, it would result fewer wasted tax dollars. Underutilized hardware can spell costly waste. For example, powering down PCs that aren't in use can result in as much as $45 in energy savings per system and $30 per monitor, per year, according to Energy Star.
"The average desktop PC currently wastes half of the power it receives," said Gov. Pawlenty in a written statement. "Having states increase the energy efficiency of their computing equipment will save consumers and taxpayers money, while reducing greenhouse gas emissions that contribute to climate change."
Under the terms of the agreement, these states have committed to reducing energy consumption from a majority of their computing equipment by:
- pledging to purchase high-efficiency computer equipment that meets or exceeds federal Energy Star ratings;
- optimizing existing computer systems by educating employees about more efficient and effective computer power-management strategies, such as better using hibernation and sleep modes;
- and ensuring in subsequent years states purchase computing equipment with increasing levels of efficiency.
Beyond preserving precious tax dollars, the initiative could spur citizens and merchants to embrace more sustainable practices themselves, observes Bill Weihl, co-chairman of the CSCI board of directors and Google's green energy czar. "[Their] actions will provide leadership for residents and local businesses looking to implement effective emission reduction policies," he said in a written statement.
CSCI, an initiative started by Google and Intel, brings together manufacturers, businesses, organizations and individuals to significantly increase the energy efficiency of computers and servers. The group is led by representatives from nine board of director companies, including Dell; Electronic Data Systems; HP; Intel; and Microsoft.
For more information about the SCEF Initiative, go to nga.org/ci. For more information about CSCI, visit climatesaverscomputing.org.
Posted by Ted Samson on November 7, 2007 10:24 AM
November 06, 2007 | Comments: (0)
IBM aims at trimming electric bills with Active Energy Manager
Management tools aimed at reining in datacenter energy waste -- and associated high costs -- continue to hit the market. The latest example comes from IBM in a software package called Systems Director Active Energy Manager (AEM).
Products such as AEM, as well as HP's Systems Insight Manager and Cassatt's Active Response, share similar features such as power capping. This function enables admins to reduce the amount of energy a given server consumes. The rationale is, servers -- particularly underutilized machines -- generally don't require as much juice as recommended out of the box. Thus, they end up using more electricity and generating more heat than necessary.
In addition to the power-capping capabilities, AEM lets admins monitor energy-usage trends to help in planning new workloads throughout the datacenter. That's of particular use to admins grappling with tight power budgets.
"Active Energy Manager gives clients a way to understand exactly how much power is being used in their data centers and where it is being consumed," said Rich Lechner, IBM's vice president for IT optimization, in a written statement. "Along with being able to cap the energy that powers these systems, this information can help save significant energy and cooling costs and create a greener datacenter environment."
Originally introduced in 2005 at PowerExecutive, AEM manages power usage across supported servers through functions such as power capping and a power-savings mode, which IBM says can potentially save datacenter operators as much as 30 percent of system power demand.
The power capping lets users set a maximum power level per system; power-savings mode lets users manage power usage as work activity shifts across various demands. This mode is enabled via an on/off switch, which can be scheduled during times of low utilization, IBM says. On supported machines, the function can be automated based on processor utilization. In addition, alerts and messages can be sent to higher-level systems management applications to automate the process.
In addition, AEM includes power-trending and thermal-trending features to monitor and report system energy usage as well as inlet and exhaust air temperatures for individual systems. "The offering allows finite and localized temperature adjustments within the IT shop to further reduce energy costs for cooling," according to IBM.
IBM says that AEM exploits iPDUs (intelligent power distribution units) to support older servers and low- and mid-range storage devices. By plugging these systems into a supported iPDU or a smart power strip, AEM collects power information and presents it in a centralized place, according to the company.
Originally developed for IBM's x86 System x hardware, AEM supports additional IBM's Power Systems and System Storage platforms. Big Blue plans to extend support to IBM System z mainframes. In addition, the company says "some hardware from other manufacturers is supported."
AEM also provides energy management data used by Tivoli enterprise solutions such as IBM Tivoli Monitoring and IBM Tivoli Usage and Accounting Manager.
IBM Systems Director Active Energy Manager will be available for download beginning Dec. 7. Pricing will start at less than $100 per system and include both Power Savings Mode and Power Capping. iPDU capabilities, power trending, and thermal trending are free AEM features.
Posted by Ted Samson on November 6, 2007 02:44 PM
November 05, 2007 | Comments: (0)
Cassatt unveils Active Reponse suite for cutting datacenter energy waste
Company says app-aware, platform-agnostic software intelligently powers servers off and on as needed
With backing from big-name IT companies such as Sun, Citrix, and Red Hat, Cassatt today unveiled its Active Response 5.0 software suite, capable of actively powering servers on and off as needed and pooling physical and virtual server resources to reduce datacenter energy waste by as much as 50 percent, according to the company.
Combining Cassatt Active Power Management and utility computing technologies based on Cassatt Collage, Active Response is designed to rein in the amount of power wasted by servers that often sit idle for long periods until they are needed. Those include machines in development and test centers, scale-out production environments, failover, and disaster recovery sites.
Specifically, Active Response employs a policy engine that, according to Cassatt, powers servers up and down on an as-needed basis, "[taking] into account server priorities and needs, server interrelationships, application dependencies and relationships, and individual server power consumption."
The software also takes into account variables such as peak and off-peak power schedules, time of day, and emergencies such as "demand curtailment" mandates from power companies to reduce electrical consumption.
Admins also can set policies to ensure server resources are available to meet service level agreements for application delivery.
The company says the system is application aware, knowing when and how apps must be shut down and brought back up. Further, it's aware of application interdependencies shared across multiple servers, Cassatt reports.
The company says the suite will run on any platform and works with any virtual or physical server , (including blade servers). Changes to existing hardware and software configs aren't necessary, either.
All told, there will be four editions of the Active Reponse suite: Standard, Premium, Data Center, and Enterprise. Especially interesting to me is the Enterprise Edition, slated for release in next year. The company says it will enable an organization to share and optimize its resources across multiple data centers in different geographies.
"Pooling your IT resources and then using software to intelligently and actively control them can create a real shift in the economics of running a datacenter," said Michelle Bailey, research vice president for IDC, in a written statement. "With products like Cassatt Active Response, customers can start to see the type of energy efficiency, application availability, and datacenter responsiveness that previously were only possible by having much higher levels of infrastructure and operations investment."
Cassatt, launched in 2003 and based in San Jose, Calif., has announced a fairly impressive array of supporters for its technology, including Sun, Citrix, Red Hat, and BearingPoint.
"We view green initiatives that promise to reduce energy with minimal resources as a win-win for everyone, and as a platform with the potential to transform the data center," said Frederic Veron, managing director, financial services infrastructure solutions at BearingPoint, in a written statement. "We are very pleased to be working with Cassatt on Active Power Management solutions that will enable our clients' data centers to intelligently power up or down in synch with business cycles and demand."
Cassatt Active Response 5.0 will ultimately be available in four editions. Standard Edition, out on Nov. 16, starts at $200 per managed machine. The Premium Edition, available today, starts at $1,250 per managed machine. The Data Center Edition, also available today, starts at $2,500 per managed machine. Finally, the Enterprise Edition will be available in the second half of 2008.
Posted by Ted Samson on November 5, 2007 12:45 PM
October 22, 2007 | Comments: (0)
Citrix unveils energy-saving PowerSmarts for Presentation Server
During peak hours, admins want their servers fully charged and running at full bore to ensure their apps are delivered without annoying service lapses. But during non-peak hours, there's no logical reason why machines need to feed on as many watts of energy.
In an effort to help cut down on that energy waste -- and the associated energy and cooling bills -- Citrix today announced a new PowerSmarts feature for its Citrix Presentation Server line, which lets admins set policies to "automatically dial server power down based on application traffic levels," according to the company.
The company claims the feature can help customers datacenter energy consumption costs by up to fifty percent, a rather impressive claim if it's accurate.
"The key to any effective response to green IT challenges is to be able to maintain, and even increase, the capacity of IT infrastructure while reducing its energy consumption," said Scott Herren, Citrix group vice president and general manager of the Citrix Application Virtualization Group. "By powering down unused Presentation Servers during off-peak usage, we are helping our customers be more green and at the same time putting money right back into their wallets."
Citrix also announced that HP is the first company with hardware to support the PowerSmarts feature in its HP ProLiant portfolio.
Notably, HP itself unveiled a power-capping feature of its own earlier this year called HP Insight Power Manager, part of the company's Systems Insight Manager (SIM) hardware management platform.
PowerSmart is compatible with all current versions and editions of Citrix Presentation Server and iLO-enabled HP servers. It will be available in December to all Citrix Presentation Server customers as a free download at www.citrix.com/cdn.
Posted by Ted Samson on October 22, 2007 02:56 PM
October 20, 2007 | Comments: (0)
Microsoft to trim Windows bloat for 2010
I continue to be interested in how app and OS code plays in to the energy-efficiency equation. What it seems to boil down to is, the smaller the application footprint, and thus the less memory and CPU power it needs, the lower the power requirements of the hardware.
With features such as the Aero UI, Microsoft Vista has been deemed a pretty hardware-intensive OS, especially if users want to reap every shiny bell and whistle it has to offer. I was thus interested by an IDGNS article from the other day about Microsoft's plan to slim down the Windows kernel.
Microsoft's Distinguished Engineer Eric Traut had this to say about Windows: "A lot of people think of Windows as this large, bloated operating system, and that's maybe a fair characterization, I have to admit," Traut said. "But at its core, the kernel, and the components that make up the very core of the operating system, is actually pretty streamlined."
The company has created a beta version of the OS, called MinWin, which will be a part of Windows 7, and which lacks a GUI. According to the IDGNS report, MinWin "takes up just 25MB when stored on disk, compared to the massive 4GB the full Windows Vista OS needs."
Given the focus companies have on energy efficiency these days, it makes perfect sense for MS to put the kernel -- as well as its other OS offerings -- on diets.
Posted by Ted Samson on October 20, 2007 04:16 PM
August 23, 2007 | Comments: (0)
A tour of HP's test lab datacenter reveals the company's Dynamic Smart Cooling initiative
Call me an environmentalist. Call me frugal. Or call me a glutton for punishment. I don't like turning on the air conditioning in my house here in Sacramento unless I absolutely have to (when the temperatures hit the 100s, for instance, as they did earlier this summer). My preferred method to beat the heat: Depending on what room I'm in, I'll turn on the fan that happens to be pointed in my general direction. It often gets the job done, and it's less expensive than cranking up the A/C to full blast.
The average server rack, however, doesn't have the luxury of flipping on the nearest cooling apparatus if it's getting too hot. So the traditional practice has been for datacenter operators to crank up the CRAC (computer room air conditioning) to the point where it feels like a meat locker.
That practice ensures that the hottest-running machines in the joint don't combust in a fiery explosion of hardware parts and mission-critical data. And even with best practices in place, that blanket-of-cold approach is wasteful from a "dear Lord, look at this month's energy bill" perspective.
But there's been some evolution in CRAC technology aimed at easing the pain. HP, for example, has been busily building on its Dynamic Smart Cooling (DSC) technology to help datacenter operators more efficiently chill their hardware on a more granular level, an approach the company says can deliver 20- to 45-percent energy cost savings. That could mean a cool million, depending on how large your facility is. And I had a chance to see DSC in action recently as I took a guided video tour of the datacenter at HP's test labs in Palo Alto, Calif., led by HP Fellow Chandrakant Patel, one of the DSC developers.
On the tour, I got to see the rows of server racks in HP's datacenter, all busily humming away -- yet the facility itself was surprisingly warm. (Patel likened it to summer in San Francisco, but really, I think it was warmer than that.) That's the magic of DSC: Affixed to every rack were small, black DSC sensors, which collected air-temperature measurements in real time and delivered them to the DSC's control node. In response to the readings, the system adjusts cooling, not of the entire facility, but rather just the area of the datacenter where a rack is running especially hot. And after that rack cools down, the CRAC unit for that region adjusts accordingly.
One of the cool tricks with the DSC, according to Patel: You can check on the temperature of your datacenter facility from anywhere, which is a mixed blessing if you're on a much-needed vacation in Tahiti and get an SMS about an overheating episode.
Patel also told me about HP's recently unveiled Thermal Assessment Services (TAS). Through TAS, HP measures a datacenter's thermal conditions to assist customers with planning server-rack placement. By knowing, for example, which regions of the facility get the most cool air, a datacenter admin could know the optimal location for the highest-utilized and hottest-running server racks.
Depending on how much a company is willing to shell out, a TAS assessment could include generating a "thermal zone map," a three-dimensional model depicting how much and where datacenter air conditioners are cooling.
As HP describes it, the maps can help datacenter operators see, for example, where there's over-provisioning or redundancy in cooling coverage in the room.
I don't want to give away everything in the video, but I will add that Patel shared with me an interesting chip-cooling technology that HP is working on called ink-jet cooling. It borrows from the technology behind ink-jet printers, but rather than showering paper with ink, small components within servers would shower chips with coolant, as needed. Time will tell just how effective this will be. I wonder about adding yet another delicate part to a server that can break.
Anyway, enjoy the video. Patel's a very engaging, eloquent, and erudite fellow.
Posted by Ted Samson on August 23, 2007 03:00 AM
August 22, 2007 | Comments: (0)
Sun celebrates green datacenter innovations and tools
As part of its Eco Innovation Initiative, Sun shined a green-hued spotlight on its massive datacenter consolidation efforts yesterday and unveiled a host of tools and services to help datacenter operators inject some energy-efficiency and eco-friendliness into their own facilities.
At a high level, Sun's boasting significant benefits from the massive undertaking, most of which affected its datacenters in Santa Clara, Calif., but also its Blackwater, UK, and Bangalore, India sites.
In Phase One of the project at its Santa Clara site, Sun consolidated 2,177 servers down to 1,240. For example, they replaced 88 Sun Fire V880 systems with 58 Sun Fire T2000 and T1000 systems yielding a 91 percent reduction in datacenter floor space and a 60 percent reduction in power costs.
The company also consolidated 738 storage devices down to 225; and 550 racks down to 65 (reaping 88 percent compression of square footage). Power-wise, the company is shrinking consumption from 2.2MW to 500KW, which will result in an estimated $1.1 million cost savings per year.
In Phase Two, which entailed a new datacenter design, build-out and deployment, the company went from 254,000 square feet of facility space to 127,000, and it expects to reap another 30 percent in energy savings.
Sweetening the effort: Sun's enjoying nearly $1 million in rebates and awards from Silicon Valley Power, including a one-time $250,000 cooling innovation award, the first award of its kind given by the utility.
Sun estimates that the company's datacenter efforts will reduce its CO2 production by 4,100 tons per year, trimming 1 percent from the company's total carbon footprint. (The latter stat is pretty interesting; most companies don't get quite so detailed as to the size of their footprint.)
One of the most intriguing aspects of Sun's new datacenter design is its pods: modular, scalable clusters of racks or benches that have the same requirements. The design will enable Sun to easily and quickly swap in and out racks, as well as deploy modular power, cooling, cabling, and monitoring equipment. Planning for future growth, Sun designed server racks to support a capacity of up to 30kw per rack.
There's a lot more detail about the modular design and other technologies and practices Sun employed on the company's Web site. I recommend that you watch Sun's Energy Efficient Datacenter Tour on the company Web site. It's pretty interesting and informative, and not too market-y, either.
In addition to showing off its shiny new datacenter facilities, Sun unveiled several products and services to help companies wring more energy efficiency out of their own datacenters.
Among them, Sun unveiled three Eco Ready Kits: The Sun Eco Assessment Kit, The Sun Eco Optimization Kit and the The Sun Eco Virtualization Kit.
The Sun Eco Assessment Kit "provides a methodical approach to analyzing datacenter energy efficiency, using a combination of assessment services for systems, storage, and datacenter infrastructure."
The Sun Eco Optimization Kit is designed "to help customers optimize, consolidate, refresh, and recycle their hardware infrastructure ... ."
The Sun Eco Virtualization Kit "offers virtualization solutions that enable better asset utilization and datacenter energy efficiency," according to Sun.
Additionally, Sun announced its Eco Services Suite, which encompasses four offerings:
- The Sun Eco Assessment Service for Datacenter, Basic, which is intended to help customers maximize power and cooling efficiency in the IT infrastructure running Web-based services;
- The Sun Eco Assessment Service for Datacenter, Advanced, "a comprehensive datacenter service providing a technical evaluation of datacenter energy use, cooling capacity, rack placement, air distribution and other environmental factors";
- The Sun Eco Cooling Efficiency Service for Datacenter, aimed at helping companies "recover misused air conditioning capacity and direct it to the areas where it is needed";
- and the Sun Eco Optimization Service for Datacenter, through which Sun provides direct assistance with implementation of corrective actions outlined in the Eco Assessment Service.
Posted by Ted Samson on August 22, 2007 12:38 PM
August 09, 2007 | Comments: (0)
EPA: 50 ways to green your datacenter
Tucked away on pages 54 and 55 of the EPA's recently released report on server and datacenter efficiency is a rather impressive and comprehensive list of strategies for reducing energy costs in your datacenter facilities.
While many action items are certainly not as simple as, say, adjusting the thermostat a bit or plugging holes in the datacenter floor, it would behoove any datacenter operator to run down the list and at least do some research on those items that seem feasible or at least worth contemplating.
One of the more intriguing proposals, to me, is at the top of the list: "Design software to avoid excess code and inefficiencies." The topic of software in the context of energy efficiency is one that I hope to explore in the near future. I can't help put wonder if it contributes to the increasing interest that we're seeing in Linux in the datacenter.
Following is the list of "Potential Energy-Efficiency Improvement Opportunities for Servers and Data Centers" as outlined by the EPA. (I've sprinkled in some links to InfoWorld coverage for some of the topics.)
Computing software
1. Design software to avoid excess code and inefficiencies (treat CPU cycles as a finite resource)
2. Provide developer tools to help improve efficiency of software
3. Enable shifting of computational load among systems for maximizing energy efficiency
4. Upgrade applications no longer supported on latest technology and/or operating systems, allowing removal of legacy servers
5. Implement virtualization to allow consolidation of server and storage hardware
IT hardware (computing, storage and network)
Operational Improvements
1. Turn off (ideally remove) dead, obsolete, or excess equipment
2. Turn off or power-manage equipment that won't be used for extended periods of time (e.g., development systems not in active use, systems for future expected increases in activity, etc.)
3. Enable power-management features on existing equipment (e.g., frequency voltage scaling)
4. Maximize utilization of storage capacity through shared data storage, data compression, and data de-duplication
Design improvements
1. Accept high-efficiency power supplies over full operating range (including DC-DC conversions) or directly accept moderate DC voltage
2. Digitally control power supplies to better match output to load
3. Use high-efficiency variable speed fans (within IT equipment)
4. Reduce energy use at lower utilizations (whether the resource is processing capacity, memory, communications, or etc.). Applies to individual systems and to clusters.
5. Improve microprocessors to lower leakage current, increase system integration, etc.
6. Use storage virtualization and massive array of idle disks (MAID) technologies to allow storage power management
7. Use centralized servers (large systems) to improve sharing of computer resources
8. Improve hardware support for virtualization
9. Use built-in power monitoring
Electrical Systems
1. Use high-efficiency power distribution (i.e., higher-voltage AC or moderate-voltage DC (50-600 VDC))
2. Use premium-efficiency motors in fans and pumps
3. Use high-efficiency UPS units over full range of load
4. Use rotary-based UPS units
5. Right size power distribution and conversion to optimize efficiency
6. Use on-site generation with grid as back-up
Heat Removal
1. Improve airflow management (i.e., use hot/cold aisle configuration or penetration sealing)
2. Adjust environmental conditions (temperature and humidity set points) to allow wider range while still meeting manufacturer specifications
3. Optimize data center airflow configuration using visualization tools (computational fluid dynamics modeling or infrared tomography)
4. Use high-efficiency variable-speed air-handler fans and chilled water pumps
5. Use variable-speed chillers
6. Use variable-speed, primary-only chilled water pumping
7. Use high-efficiency chiller and chilled water supply motors
8. Use high-efficiency CRAC units
9. Use air-side economizers (outdoor air) when outdoor conditions permit (preferably in mild climate locations)
10. Use water-side economizers (cooling tower) when outdoor conditions permit (preferably in mild climate locations)
11. Commission infrastructure systems to ensure set points are at proper values, sensors are in calibration, airflow is within design tolerances, etc.
12. Rebalance air-handler system after significant IT reconfiguration
13. Size systems and configure redundancy to maximize efficiency e.g., use redundant air-handler capacity in normal
14. Increase chilled water supply and return temperature difference to reduce chilled water flow
15. Optimize chilled water plant (cooling tower) Reuse waste heat for space heating
16. Use direct liquid cooling (water or other dielectric liquid) with currently available technology (i.e., in-rack or in-row cooling) and emerging technology (i.e., in-chassis or chip-level)
Controls and Management
1. Use system management hardware/software that enables powering down (to sleep and/or off) parts of server clusters during times of low utilization
2. Dispatch non-time-sensitive computational operations to reduce peak computing load and allow reduction of total
3. Provide for standard reporting of power use, platform temperature, and processor utilizations to assist operators in understanding and managing energy use in their data centers
4. Use shared computing models, such as grid computing
5. Optimize cooling controls to dynamically match the cooling supply to the IT heat load
6. Dynamically optimize the assignment of work across the data center to ensure maximum efficiency
7. Monitor power in real time
Distributed Generation
1. Use combined heat and power
2. Use renewable energy (e.g., photovoltaic panels)
3. Use fuel cells
Posted by Ted Samson on August 9, 2007 04:35 PM
August 06, 2007 | Comments: (0)
In AMD-Intel square-off, memory proves key
In a follow-up gauntlet of tests pitting Intel Xeon processors against AMD Opterons, independent research company Neal Nelson and Associates found the Opteron generally -- but not always -- delivers superior raw energy efficiency. Interestingly, server memory proved a key factor.
Nelson tested the Opteron against the Xeon last month and determined AMD to be, hands down, the more power-efficient chip. His tests and results sparked plenty of discussion, including in my blog, prompting him to perform follow-up tests.
The most significant difference between this and the previous test is that he used an Opteron 2222 CPU rather than the 8222. He stuck with the Xeon 5160. In his tests, he once again compared the energy efficiency of a couple of similarly equipped servers, both configured with 4GB and 8GB of memory, as they processed Web transactions at a variety of load levels.
When machines were configured with 4GB of main memory, the Intel Xeon-based server proved between 1.4 and 5.1 percent more efficient. However, at 8GB, the Opteron server was between 6.1 and 12.7 percent more power efficient.
"It appears that when Intel chips are installed in Intel motherboards and sold as Intel servers, the Intel claim of superior power efficiency is not supported by the empirical data," said Nelson in a written statement.
Notably, the AMD server equipped with 4GB of memory proved 33.3 percent more power-efficient in idle than the Intel; with 8GB, it was 43.4 percent more so.
Power consumption is relevant in idle mode, Nelson notes in a release about this round of tests, "since many servers spend most of their time waiting for work." He cites Robert Frances Group's finding that the average utilization of most processors in a datacenter is between 15 and 20 percent.
In the previous test, the AMD-based server consumed 7.3 to 15.2 percent less power at five different user load levels and 44.1 percent less power while the systems were idle and waiting for work, Nelson reported.
Nelson used the same testing benchmarks in this test as he did in the previous one: He processed a series of Web transactions on both servers, which were running Suse Linux Enterprise Server from Novell, Apache2 Web server software, and MySQL relational database.
However, in this series of tests, he used an Opteron 2222 instead of the 8222 as in the previous matchup. "AMD offers a '2 socket' version of the Opteron as a model 2222. It has the same cache and frequency specifications as the model 8222, but the 2222 is limited to motherboards with a maximum of 2 sockets, and it is less expensive than the model 8222," according to Nelson's test report.
The performance differences between the two were negligible; generally, the 8222 was around 1.5 percent more efficient than the 2222.
In terms of cost, Nelson estimates that the 8GB configuration of the Xeon server would sell for about $4,651; the Opteron system would cost approximately $4,252. With 4GB of memory, the Xeon machine costs around $4,277 and the Opteron, $3,961.
For more information or to download a copy of the white paper outlining the test and results, go to worlds-fastest.com.
Posted by Ted Samson on August 6, 2007 01:37 PM
August 02, 2007 | Comments: (0)
Employees keen on green but need reminding
Through clearly communicated policies, companies can sway eco-conscious workers to reduce costly waste at the office
73 percent of American workers would like to see their companies be more environmentally and socially responsible, but many aren't embracing energy-conservation practices in the office to the degree they do at home.
That's one of the take-aways from a poll recently commissioned by Harris Interactive and sponsored by Sun. Among examples of inconsistent behaviors exhibited by employees:
- 92 percent of participants said they turn of their lights at home, whereas only 52 percent of workers turn off the light when they leave a room at work.
- 58 percent said they turn off computers at home when they are done using them, whereas only 34 percent does so at work.
- 57 percent of workers are using sleep mode for their home computers, but only 44 percent do so at work.
All that adds up to a lot of waste for your company. As I've noted before, the costs alone of leaving computers on 24/7 can rack up -- as much at $75 in waste electricity per PC and monitor, per year, according to Energy Star.
And, of course, there's the environmental impact to that wasted energy. According to Sun, "if the percentage of America's 81.1 million office workers who are not turning off lights and computers at work took their energy-conscious behaviors to work with them, those workers could achieve the equivalent of taking 6.1 million cars off the road in aggregate CO2 emissions."
The problem is pretty clear, and there are some good solutions, too. Dave Douglas, Sun's VP of eco-responsibility, suggests that companies communicate to employees ways to be better environmental stewards at the office.
"If you're an employee, these survey results show that you have an enormous opportunity to make a difference for our planet and to help America's business make better use of their economic resources," Douglas said in a written statement. "If you're an employer, bridge this gap by communicating about the issue and how everyone can benefit by taking small steps. The good news is that your employees care about the environment, but they may need your help in realizing how much they can actually make a difference."
That strikes me as sound advice indeed. It's entirely possible that employees don't know whether or not they should be turning off their systems, perhaps thinking they need to be patched at night. Or, perhaps your company's current policy is to leave systems on for just that reason -- even though there are products on the market designed to wake up systems for that kind of maintenance, then put them back to sleep.
Whatever the reason for employees' lapses, it certainly makes abundant sense to send out an all-company missive explaining how workers might save energy at work. Visual reminders, such as small signs over light switches saying "Please turn off the lights when you are done with the conference room" would also help.
On a similar note, companies might consider reminding employees of the costs of printer waste. Research suggests that each end-user wastes about $85 worth of printer paper and ink each year through unnecessary prints, such as spill-over pages containing a superfluous line of text or gridlines, according to GreenPrint. (There's a great list of tips for reducing printer waste here at SeekingAlpha.)
And again, a small sign at the printer and copy station asking employees to only print and copy what they need would serve to put a dent in the pile of waste.
Posted by Ted Samson on August 2, 2007 12:29 PM
August 02, 2007 | Comments: (0)
Is IBM's Big Iron also Big Green?
As part of Project Big Green, Big Blue is pushing its System z mainframes as an energy-efficient alternative to server racks
As the list of arguably green technologies continues to expand, it's interesting to observe how many of the candidates are by no means new. Virtualization has technically been around for years, for example, as have thin clients. But one of the more surprising technologies to make the list is the mainframe.
Yes, mainframes. Big Iron. Those big ol' data-crunching behemoths regarded by some as the dinosaurs of the datacenter. Pundits have periodically predicted they'll go extinct -- or at least pondered aloud how long they'd be around.
But mainframes have continued to evolve -- and seemingly thrive. According to both IBM and IDC research, mainframes sales remain strong. And with green fever infecting the business world, Big Blue hopes that Big Iron will soon be regarded as Big Green Machinery.
As part of its billion-dollar Project Big Green endeavor, IBM revealed this week that it's partaking of a giant bowl of its own dog food, moving the workload of 3,900 of its 8,500 servers to 30 virtualized System z9 mainframes running Linux. (Yes, Big Blue says it will properly recycle those machines.)
"The cost of energy, power to run computers, storage, and networking equipment, as well as the power to the cooling equipment, is becoming the highest single cost of managing a datacenter," says David Gelardi, VP of industry solutions at IBM. "IBM took a look at these very interesting plums coming to the forefront at the same time. We have an opportunity with systems management tools, with Linux, and with virtualization, to be able to take the workloads that are principally running on much smaller, underutilized Unix servers and move them over to those 30 very large mainframes."
Big Blue anticipates the move from 3,900 servers to 30 mainframes will cut energy consumption by around 80 percent, a healthy cost savings no matter how you slice it. Specifically, the company anticipates reducing its total annual energy consumption, including power and cooling, from 3,266 kilowatts to 629, and total expenses from $2.86 million to $551,000.
Reports from Robert Frances Group (RFG) lend credence to IBM's claims that mainframes can deliver processing power more efficiently than standard servers. In a white paper titled "Mainframe Computing and Power in the Data Center" dated Feb. 16, 2007, RFG reports the following:
"Mainframe systems consume less power, both in absolute and relative terms [than standard servers]. Typically, mainframe power densities are less than half of those of current rack and blade distributed systems. When looking at like workloads, the amount of energy consumed falls precipitously, in some cases the costs associated for power needed for an application are reduced by a factor of 600."
Part of the can be attributed to the fact that "mainframes are designed with a central AC/DC power converter, which operates at over
90 percent efficiency, compared with many existing rack server power converters which operate at 70 percent," according to a separate RFG report.
Also a boon: more precious floor space, moving from 11,045 square feet of occupied space to 1,643 square feet with the 30 mainframes.
Meet the machines
IBM's z9 mainframes are 64-bit machines, packing "specialty processors" designed for processing eligible Linux, Java, and data workloads as well as encrypting and decrypting certain data.
The machines' HiperSockets technology provides fast communication among all the virtual servers contained in a single machine, according to IBM. "By contrast, in a distributed environment, where many physical servers are connected by networking cables, lag time may be greater," the company argues.
Additionally, IBM says the machines can handle massive workloads. "The mainframe recently achieved the world's largest core banking benchmark result, delivering a record 9,445 business transactions per second in real time based on more than 380 million accounts with three billion transaction histories."
"Thank God for Linux."
Green ambitions -- both in terms of ecofriendliness and slicing energy bills -- are just part of the picture here. Linux's maturity stands to boost the mainframe's appeal, Gelardi says. "If you were to talk to just about any software company in the world, they would tell you the same story: "Thank God for Linux, because the 37 Unix variants were making us crazy. Linux is attractive because it's ubiquitous. You couldn't find too many products in the industry that don't support Linux."
Also appealing, according to Gelardi: the potential savings on software licenses, which are generally sold on a per-CPU basis. Moving from 3,900 servers, which might have, on the low end, 7,800 CPUs, to the 30 mainframes, which represent, at most, 1,920 CPUs (64 per mainframes), means a substantial reduction in software bills.
Finally, IBM is making the fairly familiar case the having fewer machines means you can free up IT staff to work on other projects.
Geraldi stresses that the mainframe isn't suited for all server tasks, which is why the company isn’t trading it its remaining 4,000-plus servers for mainframes. Typically, mainframes have been used for bulk-data processing tasks such as ERP and financial transaction processing. "There are lots of workloads that will still favor other architectures. We do not today, and as far as I can tell, for the future, have this notion of this being one size fits all."
Posted by Ted Samson on August 2, 2007 03:00 AM
July 26, 2007 | Comments: (0)
Ambient Orbs spark bright idea for cutting energy waste
Clive Thompson over at Wired has shared an intriguing story about how Mark Martinez, demand response manager of Southern California Edison, got 120 customers to reduce energy consumption by 40 percent during peak periods. He gave them Ambient Orbs, small spheres that you can place on your desk and that light up and change colors in response to changing streams of data.
Thompson notes that these balls of joy were first marketed to individuals for monitoring stock performance. If the orb is blue, all is well; when red, it's time to consider quickly buying or selling.
Rather than receiving Nadsaq data, these orbs as configured by Martinez feed off data about electric rates for SCE customers -- rates that go up, of course, as demand swells. So when Martinez's orb recipients see their balls flashing red, they know that it's a good time to power down where possible (for example, adjust the thermostat, turn off excess lights, and so on).
It really is an elegant solution, as Thompson observes, providing a quiet yet constant reminder about electricity costs. It's all too easy to forget or take for granted that for every minute you have an extra light on or keep your computer and monitor running while not in use or have your servers running at full throttle when they don't need to be, you or your company is paying for it. And it's especially useful to know when demand has spiked and your rates have suddenly shot up.
So I have a suggestion for SCE and other utility companies: Build on this idea because it seems to work. But relying on Ambient Orbs, of course, isn't practical. While they're certainly neat and cool conversation pieces to boot, they're also expensive: $150 a pop (plus shipping and handling).
But how about developing a simple desktop utility that is fed the same data stream received by these orbs? When electric rates go up, a little window can pop up telling you as much -- and perhaps provide some tips as to what actions to take. I know I'd be happy to donate a little bit of my computer-processing power to run such an application.
On a related note, I have a free utility from Uniblue called LocalCooling installed on my system, and it's designed to optimize a PC's power consumption. Among its features, it displays how many killowatt hours, trees, and gallons of gasoline you've saved by managing you computer's energy usage.
(Thanks to TreeHugger for bringing this story to my attention.)
Posted by Ted Samson on July 26, 2007 02:17 PM