Sustainable IT | Ted Samson | TAG: Intel

MORE ENTRIES

TAG: Intel

January 28, 2008 | Comments: (0)

In an effort to fuel the growth of the clean-energy market, Intel today announced plans to spend an undisclosed sum purchasing more than 1.3 billion kilowatt hours per year in renewable energy credits (RECs). According to the U.S. Environmental Protection Agency, this move makes Intel the single-largest corporate purchaser of green power in the United States.

As explained by the EPA, "RECs (also known as green tags, green energy certificates, or tradable renewable certificates) represent the technology and environmental attributes of electricity generated from renewable sources. Renewable energy certificates are usually sold in 1 megawatt-hour (MWh) units; a certificate can be sold separately from the MWh of generic electricity it is associated with. This flexibility enables customers to offset a percentage of their annual electricity use with certificates generated elsewhere."

Sterling Planet, a national supplier of renewable energy, energy efficiency and low-carbon solutions, will handle Intel's REC purchase, which includes a portfolio of wind, solar, small hydro-electric and biomass sources.

"The EPA estimates that Intel's REC purchase has the equivalent environmental impact of taking more than 185,000 passenger cars off the road each year, or avoiding the amount of electricity needed to power more than 130,000 average American homes annually," according to Intel's announcement.

These green tags/RECs shouldn't be confused with white tags, or energy efficiency credits, which represent megawatt hours saved through power-saving projects such as server consolidation.

I commend Intel here for its move. In an ideal world, of course, companies seeking to run their operations on clean technology would be able to draw on clean energy straight from the grid, rather than having to buy RECs. Given the momentum of the clean-tech movement, that should become feasible in coming years.

Posted by Ted Samson on January 28, 2008 02:41 PM

January 16, 2008 | Comments: (0)

AMD yields more efficiency, Intel more throughput in quad-core CPU showdown

AMD yields more efficiency, Intel more throughput in quad-core CPU showdown In a recent gauntlet of tests comparing AMD's quad-core Opteron processor (Barcelona) to Intel's quad-core Xeon (a.k.a. Tigerton), the Xeon delivered up to 14 percent more throughput, but the Opteron used up to 41 percent less energy.

The series of tests were conducted by Neal Nelson and Associates, an independent consulting firm. If you've been following my blog for a while, that name may sound familiar: Nelson has run similar power and performance tests over the past year, one in July and one in August.

Nelson conducted this series of tests using similarly configured quad-core Xeon and Opteron servers using 1GB memory modules at 4GB, 8GB, and 16GB main memory sizes. He also used one- and two-socket configurations at speeds of both 2.0GHz and 2.33GHz.

Cutting to the chase, Nelson determined the following in his tests:

-- When the sizes of the database working sets were small enough to fit in the servers’ kernel disk buffer cache, such that there was virtually no physical disk I/O, the Xeon-based servers delivered up to 14 percent higher throughput than the Opteron-based servers.

-- When the sizes of the database working sets were too large to fit in the kernel disk buffer cache, which forced substantial physical disk I/O, the Xeon-based servers delivered up to 3 percent higher throughput than the Opteron-based servers

-- When the servers were subjected to various identical levels of transaction arrival rates, the Opteron-based servers consumed up to 32 percent less power than the Xeon-based servers.

-- When the systems were idle and waiting for transactions to process, the Opteron-based servers consumed up to 41 percent less power than the Xeon-based servers. ("The power consumption at idle is particularly significant since studies have shown that many servers are powered on, but idle, 80 percent of the time," Nelson notes.)

"By themselves, the Intel processor chips may use less power, but all current Intel Xeon servers require the use of fully-buffered memory modules [FB-DIMM]. These FB-memory modules appear to consume more power than the DDR-II memory modules used by the AMD-based servers. The result is that in many cases an Opteron-based server actually uses less total power than a Xeon-based server," says Nelson in a written statement.

A better green benchmark?
Nelson reached his conclusions by employing what's he's dubbed Neal Nelson's Power Efficiency Benchmark. The benchmark works as follows: Nelson simulates users from 32 separate computers submitting individual transactions to similarly configured servers running Apache2 Web server, the MySQL relational database, and Novell's Suse Linux Enterprise Server O. He measures the throughput and power usage of the systems in increments of 50 users, from 100 to 500, over half-hour sessions. "The benchmark has a complex multi-user load with a large memory footprint, a high volume of context switches, significant network traffic, and substantial amounts of physical disk I/O," according to Nelson.

In addition to sharing his conclusions for his test, Nelson has made an effort to differentiate his power-efficiency benchmark from the one recently unveiled by SPEC. "The SPECPower test has a single-client machine feeding batches of 1,000 transactions to a small number of Java-based application programs," says Nelson. "[It also] has a small memory footprint, a low volume of context switches, simple network traffic, and it performs no physical disk I/O. The SPEC test was created by a committee of computer vendor employees, and SPEC offers no guarantee that their numbers will correlate to a customer's real-world experiences."

Nelson's test results can be viewed in their entirety on his Web site.

Ted Samson is a senior analyst at InfoWorld and author of the Sustainable IT blog. Subscribe to his free weekly Green Tech newsletter.

Posted by Ted Samson on January 16, 2008 11:50 AM

September 13, 2007 | Comments: (0)

In the chip race, efficiency beats speed

Following is essentially the content of today's Green Tech Newsletter. I generally don't post the newsletter verbatim in my blog, but it seemed important enough in the context of my piece about Barcelona. The newsletter, by the way, is free. You can subscribe here.

AMD and Intel have been locked in fierce races for quite a while now, the latest being the race to first deliver a quad-core processor. Intel managed to win by a nose there, unleashing Tigerton a few days before AMD presented Barcelona.

But speed alone won't be the determining factor as to which of these two processors will reign in the datacenter. Driven by customer demand for more energy-efficient hardware, both companies are talking up the importance of a less-familiar metric called performance per watt (PPW). As Intel tells it, "performance per system watt is calculated by taking the performance score from a benchmark or application and dividing it by the average system power usage (AC power from the wall)."

At a press meeting with Randy Allen, VP in charge of AMD's server and workstation division, Allen confirmed that there's been a shift in what customers want from hardware. "The traditional buying criterion has been peak performance. People would buy at the highest peak performance they could or they would buy on performance per dollar. ... This emergence of performance per watt has been dramatic over the last two years."

The problem with PPW is it's difficult to measure. To draw on a familiar example, what's a better performer: a hybrid or a pickup? The hybrid will certainly deliver more miles to the gallon if you're using it to get around town for work or play, but if the task is hauling lumber or heavy equipment, you'll get better results with the truck. Then again, if speed truly is what you need, gas costs be darned, maybe that Porsche is right for you.

Back to servers, then, how does a datacenter operator go about assessing whether Machine A or Machine B will deliver better PPW for, say, the company's accounting application? In my interview with Bruce Shaw, director of server operations at AMD, he told me the best approach is to load up your app on your server, then measure it at the wall.

Sure, that's not as simple as being able to look at the vendor-provided figures and know for certain which machine is the best overall performer for your specific needs. Then again, the exercise is a valuable one, given that filling your racks with the machines that gives you the best PPW will save you money in the long run -- just like filling your corporate fleet with hybrid trucks just might make more sense economically and ecologically than a fleet of Segways.

Posted by Ted Samson on September 13, 2007 03:00 AM

August 30, 2007 | Comments: (0)

Power-efficiency test reveals respective strengths of AMD and Intel

In its latest round of power-efficiency tests pitting the AMD Opteron against the Intel Xeon, independent consulting firm Neal Nelson and Associates found that AMD's offering outperformed Intel's in 36 of 57 cases.

The results are by no means cut and dry. While the AMD appears to once again have an edge in terms of raw power effienecy, factors such as memory size, transaction type, and transcation loads made for notable and interesting differentiators.

Nelson performed this gauntlet of tests on servers -- one equipped with the Opteron 2222 and the other with the Xeon (Woodcrest) 5160 -- configured with two, four, six, and eight gigabytes of main memory at various transaction-processing load levels.

Overall, Nelson found that for certain configurations and at certain load levels, the Intel Xeon based server was 2.4 to 11.7 percent more power efficient while in other cases the AMD Opteron based server was 9.2 to 23.1 percent more power efficient.

Memory once again proved an important variable. In general, larger main memory sizes resulted in higher transaction throughput and higher power efficiency. Further, in cases where Intel outperformed AMD in power efficiency, the servers were configured with smaller larger memory sizes. "There was a visible trend that as the memory size increased that there was an increasing shift of power-efficiency toward the Opteron," Nelson notes the white paper outlining his testing.

Importantly, Nelson discovered differences in power-performance depending on what type of work the servers were doing. At the maximum throughput, based on transactions per watt hour, the Intel system delivered better power-efficiency by 5.0 to 5.5 percent for calculation intensive workloads. For disk I/O intensive workloads, AMD delivered better power efficiency by 18.4 to 18.6 percent.

In addition, when the systems were idle and waiting for transactions to process, the AMD server was 30.4 to 53.1 percent more power efficient.

He put the machines through two different tests. One employed the Neal Nelson Transaction Benchmark, in which simulated Web clients present transaction requests to the server. As soon as the server responds to a request, the client submits a new request.

In the second test, employing the Neal Nelson Power-Efficiency Benchmark, he presented the servers with a set number of transactions, then measured the power expended for each transaction arrival rate.

For the loads, he simulated over-the-Web credit card transactions on the servers from RTE (Remote Terminal Emulator) nodes to the machines, which were running Apache2.

You can read the white paper outlining the testing and results here on the Neal Nelson and Associates Web site.

Posted by Ted Samson on August 30, 2007 05:24 PM

August 14, 2007 | Comments: (0)

A mighty ionic wind could revolutionize chip-cooling

InfoWorld: A mighty ionic wind could revolutionize chip-cooling Researchers at Purdue University have demonstrated a new technology that employs "ionic wind engines" to cool chips as much as five times more effectively than other experimental cooling approaches, the institution reports.

In their testing, researchers found that the technology reduced heating from about 140 degrees Fahrenheit to about 95. This kind of cooling technology could help hardware makers develop thinner, more powerful laptops, according to Purdue. (I imagine it could also show up in server hardware.)

"In computers and electronics, power equals heat, so we need to find ways to manage the heat generated in more powerful laptops and handheld computers," said Timothy Fisher, associate professor of mechanical engineering at Purdue.

According to Purdue, the cooling technology could appear in computers in the next three years -- if researchers are able to reduce the size of components within the device from the scale of millimeters to microns, or millionths of a meter. They'll also have to figure out how to make the system sufficiently rugged.

"As things get smaller, they get more delicate, so we need to strengthen all the elements. And we believe we can achieve this goal in a year or so," said Suresh Garimella, a professor of mechanical engineering at Purdue.

Down the road, it could also show up in portable devices such as cell phones.

The technology uses tiny "ionic wind engines" to achieve its supercool chilling effect. Here's how Purdue explains it: The device contains a positively charged wire, or anode, and negatively charged electrodes, called cathodes. When voltage passes through the device, the negatively charged electrodes discharge electrons toward the positively charged anode. Along the way, the electrons collide with air molecules, producing positively charged ions, which were then attracted back toward the negatively charged electrodes. This creates the "ionic wind," which increases the airflow on the surface of the experimental chip.

The research, funded by Intel, was conducted and documented in a paper by Fisher, Garimella, mechanical engineering doctoral student David Go, and Intel research engineer Rajiv Mongia. Specifics about the research will appear in the Journal of Applied Physics, tentatively scheduled for the Sept. 1 issue.

Posted by Ted Samson on August 14, 2007 11:44 AM

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

July 20, 2007 | Comments: (0)

Study: AMD more power-efficient than Intel

AMD Opteron servers proved more energy efficient than those running Intel Xeon in a server-power-efficiency test performed by Neal Nelson and Associates.

Specifically, the independent computer-testing firm announced today that in the tests, "the AMD based server used 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."

That translates to annual electricity savings between $20.29 per server and $36.04 per server, depending on the workload, the study concluded. At idle speeds, it amounts to a $99.76 per-server, per year saving.

"AMD must have put a lot of energy into optimizing the power usage for their products and it appears that AMD's customers will now realize significant energy savings," said Neal Nelson, president of the testing group, in a written statement.

Neal Nelson and Associates took a new approach for this test, employing at client-server benchmark where Web transactions were processed against a server running Novell SUSE Linux, Apache2, and MySQL. The tests were run on similarly configured 3GHz Intel (Woodcrest) Xeon and AMD Opteron servers, according to Neal Nelson and Associates.

Following is a table showing how the test broke down:

Not surprisingly, AMD was pleased with the test results, though as I write, no one at the company has had a chance to fully review the report. "While we did not review the methodology for these tests, we are not surprised by the results as they reiterate what we hear from customers and see in our own labs," said John Fruehe, manager of worldwide market development for server/workstation products at AMD, in a written statement. "We appreciate Neal's efforts to shed additional light on energy-efficient server computing, as it further validates what we've known is an important issue for our customers for some time. AMD is committed to delivering energy-efficient solutions to our customers, as you'll see when we roll-out our native Quad-Core processor, codenamed 'Barcelona,' (with the same power and thermals as our dual-core processors) later this summer ... ."

Intel issued this comment today: "We stand by all of our energy efficient claims, period. We also recommend that IT managers who don't do their own in-house testing turn to the dozens to hundreds of independent and certifiable benchmark organizations for the best, most credible perspective."

There is a thread on the subject below, and Neal Nelson has taken some time to address the specific questions and comments some of you have raised.

Also, there are discussions going on at Slashdot and Xtreme Systems.

For those of you who are interested in reading the specs on the servers and details of the test methodology, the report can be viewed here [PDF].

Posted by Ted Samson on July 20, 2007 12:52 PM

Technology White Papers

InfoWorld Technology Marketplace

» Technology White Papers Library

Technology White Papers by Topic

Technology White Papers E-mail Alert