Green Data Center Blog

IBM claims 90% Top 20 Energy Efficient Supercomputers

IBM will release on Monday, July 13 a press release on its energy efficient supercomputers.  I don’t have the link yet, but below is the text.

Ironically, I just sat down with a person at ARM to discuss energy efficiency and he was mentioning how in 2005 he asked about energy use at a super computing conference, and people would think he was asking a stupid question.

It’s great to see the question “how much energy does your supercomputer use?” is now a normal part of a purchase decision.

Note this quote below.

"Modern supercomputers can no longer focus only on raw performance," said David Turek, vice president, deep computing, IBM. "To be commercially viable these systems most also be energy efficient. IBM has a rich history of innovation that has significantly increased energy efficiency of our systems at all levels of the system that are designed to simultaneously reduce data center costs and energy use."

Report Finds IBM Supercomputers Most Energy Efficient in the World
IBM Dominates Green500; 90 percent of Top20 Energy Efficient Supercomputers Made by IBM, Staggering 57 Percent of Top100 from IBM

ARMONK, N.Y., July 13, 2009. . . A new list announced today found that IBM (NYSE: IBM) supercomputers already deemed the most powerful in the world are also the most energy efficient according to the findings of the latest Supercomputing 'Green500 List' announced by Green500.org.

Energy efficiency--including performance per watt for the most computationally demanding workloads--is a core design principle in developing IBM systems.  IBM offers the broadest range of generally applicable supercomputers represented on the Green500 List including Blue Gene, Power servers, iDataPlex, BladeCenter and hybrid clusters.

The list shows that 18 of the Top20 most energy efficient supercomputers in the world are built on IBM high performance computing technology. The list includes supercomputers from across the globe being used for a variety of applications such as astronomy, climate prediction and pharmaceutical research. IBM also holds 57 of the Top100 positions on this list.

The number one most energy efficient system in the world -- an IBM supercomputer based on an IBM BladeCenter QS22 at the Interdisciplinary Centre for Mathematical and Computational Modeling, University of Warsaw -- produces more than 536 Mflops (millions of floating point operations per second) per watt of energy.

The world's fastest supercomputer, the IBM supercomputer at Los Alamos National Laboratories, the machine that first broke through the petaflop barrier, is ranked the fourth most energy efficient supercomputer in the world capable of over 444 Mflops per watt of energy, while the second  fastest supercomputer in the world manufactured by Cray is ranked 90th on the Green500 List, producing only 152 Mflops per watt.

"Modern supercomputers can no longer focus only on raw performance," said David Turek, vice president, deep computing, IBM. "To be commercially viable these systems most also be energy efficient. IBM has a rich history of innovation that has significantly increased energy efficiency of our systems at all levels of the system that are designed to simultaneously reduce data center costs and energy use."

The Green500 list is published by Green500.org. It provides a ranking of the most energy-efficient supercomputers in the world and serves as a complementary view to the TOP500 list of worldwide supercomputers announced last month by Top500.org.

More information about the Green500 List is available at http://www.green500.org
More information about the TOP500 List is available at http://www.top500.org
More information about IBM and HPC Solutions: www.ibm.com/deepcomputing

Instant Virtualized Physical Infrastructure – Stratascale addresses market for hybrid physical and virtual servers.

Just interviewed Reed Smith, Director of Product Management for StrataScale and discussed their IronScale product.  Their announcement is here.

This is an interesting extension of service from Raging Wire collocation services to host a virtualized infrastructure.

In many ways the Stratascale offering is a greener data center solution for companies with 50 – 500 employees who run their own servers on site or even a collocation offering. Why? Because the solution is designed with Virtualization as an assumption.  The good thing is versus services like Amazon Web Services you can also choose to have a no virtualization and be direct on hardware.  Also, the servers are not share with other customers. The virtualized servers are all yours.

This was my first chat with Reed, but I am sure I’ll be talking to him again to discuss Stratascale’s solution.

Part of Stratascale’s value is its UI for system provisioning.

You can watch a demo here.

The physical servers are listed.

Available in 3 levels of automated integrated bundled environments, IronScale is built on real, physical dual- and quad-core x86 servers.

Level 1 Server...2 cores, 4GB RAM, 70GB storage

Level 2 Server...4 cores, 8GB RAM, 70GB storage

Level 3 Server...8 cores, 16GB RAM, 70GB storage

Each server bundle features:

Dual-/quad-core Intel^® Xeon^® CPUs

70GB of local RAID 5O storage

Your choice of a Red Hat^® Linux^® or Windows^® Server OS

1 Mbps dedicated bandwidth

2 networks and 8 external IP addresses per client

100 internal IP addresses per network

VPN for 1 site-to-site and 5 remote users per client

24x7 Monitoring and Management

KVM access

The data center facility is listed is run by Raging Wire.

Our Tier IV Data Center

Staff

Multi-disciplined, certified engineering staff and 24x7 support team of IT experts is rigorously trained, has extensive industry expertise, and is committed to clients, standards, and best practices.

World-Class Facility

Tier IV class "A+" 200,000+ square foot data center

Engineered for 99.999% availability

Power and cooling is scalable beyond 200 Watts per sq. ft.

Carrier neutral high-speed Internet, over 20 Gigabits of bandwidth

N+2 minimum system and component redundancy for concurrent maintainance and fault tolerance

On-site, 69Kv power substation and well

Financial-grade physical security

Their PUE is not advertised, but given the highly virtualized environment the performance per watt should be high.

Environmental Responsibility

At RES, we have a healthy respect for our environment. Which is why we have always built common sense and green practices into everything we do. But being environmentally responsible isn't just good for the community and the planet - its good business. For our efforts, we have received numerous environmental awards, but we don't stop there. We constantly refine our processes to improve our power and resource efficiency, reduce and recycle wastes, and help us to operate more productively.

Efficiency and sustainability is key to everything we do.

Our power conservation efforts have saved more than 4,000,000 kWh of electricity.

We've increased our chilled water plant and cooling efficiency, conserving an additional 250,000 kWh of electricity per month.

Our two chemical-free water treatment systems have eliminated chemical use entirely and reduced our water discharged by 80%.

We recycle 100% of the cardboard, steel, copper, and aluminum we use for construction and other activities. In paper and cardboard recycling alone, it is equivalent to 36+ mature trees per year.

We recycle 100% of our lead acid batteries, over 700,000 pounds worth, and donate the proceeds to the United States Wolf Refuge.

We recycle more than 2,000 pounds of electronic waste per year, from both our own company and as a free service to our valued clients.

Our social and environmental responsibility is an essential part of contributing to and protecting the community we live in.

HP’s First ENERY STAR Servers

TheRegister has an article about HP and Dell’s ENERGY STAR Servers.

HP and Dell claim energy efficiency server firsts

Power test pass with a catch

By Timothy Prickett Morgan • Get more from this author

Posted in Servers, 6th July 2009 20:40 GMT

Getting IT vendors to agree to any standard, even one that they have a big hand in shaping, is almost impossible. And so it is with the new Energy Star specification for servers, supposedly embraced by server makers to show the energy efficiency of their metal.

The Energy Star specification for servers, which debuted in May, is designed to gauge server power consumption so IT managers can shop for the most energy-efficient machines using an agreed yardstick. In this case, a yardstick established by the US Environmental Protection Agency's popular Energy Star program.

HP has their own site for ENERGY STAR servers.

 

Part of the ENERGY STAR specification are data sheets.  Here are a few HP has published.

HP ProLiant DL360 G6

• Combining concentrated 1U compute power, integrated Lights-Out management, and essential fault tolerance, the HP ProLiant DL360 G6 is optimized for space constrained installations.
  • Power and Performance Data Sheet (504633-xx1)

HP ProLiant DL380 G6

• The HP ProLiant DL380 G6 Server continues to deliver on its heritage of engineering excellence with increased flexibility and performance, enterprise-class uptime and manageability, 2-socket Intel Xeon performance, and 2U density for a variety of applications.
  • Power and Performance Data Sheet (491316-xx1)
  • Power and Performance Data Sheet (491315-xx1)
  • Power and Performance Data Sheet (504633-xx1)
  • Power and Performance Data Sheet (491335-xx1)

Access Power, Thermal and Utilization Parameters

• Windows Commands
• Linux Commands

Facebook Technical Operations VP Jonathan Heiliger tells Server OEMs “You guys just don’t get it,”

GigaOm has an interview with Facebook VP Jonathan Heileger.

Heiliger had strong words for OEMs and system builders during his chat with Om. To compete with sites like Facebook and Google, Heiliger said, OEMs and system builders need to be more power- and cost-efficient. “You guys just don’t get it,” he said, adding that Facebook has reaped success from investing heavily in its infrastructure.

InternetNews focused on the Intel and AMD dig.

Intel, AMD Get Thumbs Down from Facebook

Head of the social network's technical operations say the latest and greatest from Intel and AMD don't make the grade.

June 25, 2009
By Andy Patrizio: More stories by this author:

SAN FRANCISCO – Intel and AMD just got a nasty smackdown from the person who run's Facebook's datacenters, saying the chips don't deliver on their promises.

Jonathan Heiliger, vice president of technical operations for Facebook, was being interview by GigaOm Network founder Om Malik here at the GigaOm Structure 09 conference. Malik asked him about unexpected problems in managing the fast-growing company's datacenters.

Reps from Intel (NASDQ: INTC) and AMD (NYSE: AMD), both running panels and present at the show, must have clenched their teeth when they heard this:

"The biggest thing … was less-than-anticipated performance gains from new microarchitectures, so new CPUs from guys like Intel and AMD. The performance gains they're touting in the press, we're not seeing in our applications," Heiliger told the audience.

He didn't let the tier one server vendors off any easier. "I'm not sure if I'm embarrassed or pleased for OEM vendors in the audience, but you guys don't get it. To build servers for a company like Facebook and Amazon, and other people who are operating fairly homogeneous applications, they have to be cheap and super power efficient," said Heiliger.

He added he's not sure why hardware vendors fail at the job, but thinks customers need to step up and apply pressure. ""Perhaps in the coming months we'll see more collaboration with people running small clusters and large clusters," he said.

Netbook, Small and Disruptive – Always Connected Servers and Devices

Economist has an article on netbooks titled “Small and Disruptive.”

Small but disruptive

Jun 11th 2009 | SAN FRANCISCO AND TAIPEI
From The Economist print edition

Laptops are evolving—and forcing the rest of the computer industry to change

Imaginechina

IT WAS like waiting for Godot: in the end, the great man did not come. The crowd at Apple’s jamboree in San Francisco this week was visibly disappointed when Steve Jobs, the computer-maker’s legendary chief executive, did not even put in a brief appearance after a six-month medical leave. But another no-show was perhaps more important. Proving many techno-pundits wrong, Apple did not present a “tablet”—a pared-down computer in both size and abilities, with a touch screen. Had it done so, it might have helped settle a question that has preoccupied the personal computer (PC) industry for some time: are netbooks—cheap and basic laptops that are flying off the shelves—just a fad, or the future?

The answer is probably both, as an “iPad”, or whatever Apple’s device may be called, would have demonstrated. Netbooks are already being supplanted by a plethora of new gadgets, including tablets and increasingly computer-like mobile phones (see article). But the idea they embody, that a near-permanent connection to the internet permits simpler technology, is already changing the economics of the PC business.

As companies build low power devices that can be on all the time, connected to servers on all the time there is a disruption in the industry. Energy efficient devices that can be left on all the time are changing how people use computers and what the expect.  It is no longer the common case of I am going to sit at my PC to use Word or Excel.

Yet although netbooks have acquired many frills and mutated into new forms, the theory behind them endures: computers do not need to be stuffed with the latest whizz-bang technology if they have a high-speed connection to the “cloud” of services available online. At Computex firms showed devices equipped with WiMAX, a new wireless technology that allows for fast, ubiquitous wireless connections.

It is this combination of connectivity and cloud computing that makes netbooks and their successors so disruptive. Some mobile-network operators now throw in free netbooks if subscribers sign up for a mobile-broadband contract. This will put further pressure on prices, since mobile operators have more bargaining power than individual consumers, although it also opens a huge new distribution channel for computer-makers.

I recently participated in a NDA video shoot for a new product 6 months out which fits in this category, and it got me thinking more on this subject.

So, I placed my order for an iPhone 3G S which I’ll get on June 19..  My Apple friends will be happy to see after all the years at Microsoft, I’ll be back to using an Apple product.

HP Proliant Z6000 G6, 28% Power Savings, 31% lighter - Higher Performance per Pound

When you look at HP’s new scale out servers ProLiant Z6000 G6.

you don’t see any fans or power supply. 

The module plugs into the power and cooling infrastructure

In this fact sheet.

Delivering unrestricted airflow through a shared power and cooling infrastructure in a 2U chassis lessens fan load, resulting in a 28 percent power decrease per server compared to traditional rack-based servers.(1)

and HP put the server on a scale

Removing up to 31 percent of weight per server and up to 838.5 tons of data center weight,(2) while reducing shipping as well as construction costs with a standardized, shared infrastructure.

the cabling’s been moved to front to improve access and airflow.

Reducing service and maintenance costs with front system cabling that allows easy access to key system components.

There are power sensors.

And air pressure sensors.

HP High Efficiency Cloud Infrastructure Servers, Moving Away from Blades, Learning from Data Center Operation

Notice how IBM, Dell, Rackable, and Dell’s high efficiency cloud infrastructure servers are not blade enclosures?  Maybe it’s because HP’s data center design group who creates solutions like this

are figuring out how inefficient blades are for Cloud Computing. Quoting HP’s press release.

With the HP ProLiant SL portfolio, customers can cut acquisition costs by 10 percent and power draw by 28 percent, while doubling their compute density.⁽²⁾

“Customers with scale-out business models need solutions that make every dollar, watt and square foot in the data center count,” said Christine Reischl, senior vice president and general manager, Industry Standard Servers, HP. “The HP ProLiant SL offers pioneering customers like these the most significant design innovation since the blade form factor, allowing them to achieve an economy of scale never before possible.”

And, HP includes an IDC quote.

“Businesses built on extreme scale-out environments, such as cloud, Web  2.0 and HPC, operate at maximum transaction volume and low margins,” said Michelle Bailey, research vice president, IDC. “These customers have very distinct and unique data center requirements, specifically around energy efficiency, cost and time to market. The introduction of technology solutions such as the ExSO portfolio from HP are specifically addressing customer requirements for optimizing capitol expenditures while lowering ongoing operating costs. As a result, these solutions are helping to redefine data center economics.”

The problem with blades is high density computing created hot spots with problems airflows.  But, this behavior to use blades was driven by chargeback models that used rackspace occupied.  Which artificially can bring down IT costs when in reality it increases costs. Just read the above quotes again, on how these latest servers are the most efficient.

Note in this picture how the removal of a Proliant SL is similiar to a what a blade removal picture would be.

One great thing about Sun, HP, Dell, and IBM all getting into the data center design business is these companies are learning what the impact is  of their hardware.

Wow think about this. For the first time many Server OEMs are building data centers to host their hardware, and they need to build hardware that works best in their data centers. Whoever can create the most efficient systems has a competitive advantage.

This is what Google does.

There is a new competition in data centers, and has a higher probability of a green data center.

Hey this will make a great white paper/presentation.

Little and Large, Extremes of Nehalem-EX vs. Via Nano, the Struggle for Power Efficiency

Just posted this blog entry with Adam Bogobowicz.

Little and Large

When you hear the phrase “green IT”, you know that the word virtualization will usually appear shortly afterwards. Virtualization provides the ability to consolidate your IT infrastructure from several physical computers (each of which has a basic power overhead) onto virtual computers running on significantly fewer host computers. The increase in power consumption for each virtual computer you run on the host is considerably less than the power overhead for each replaced physical computer.

At this point, you can usually hear the sounds of hardware manufacturers rubbing their hands with glee. This enthusiasm results from the prevalent association that virtualization (and especially consolidation) requires multi-core processing capacity with associated power consumption rates. A good example of this type of hardware would be the recently announced Nehalem-EX processor from Intel. Everything is larger and more numerous on this chip when compared to the previous generation:

• 8 cores
• 9 times the memory bandwidth
• 16 simultaneous threads
• 24MB of cache
• 2,300,000,000 transistors

If consolidation is the route you want to take and you have multiple physical computers to consolidate, then this processor would certainly be a good choice. But what if you see virtualization as an aid to manageability rather than consolidation and you want to conserve power?

O Fortuna

At the opposite end of the spectrum from Nehalem-EX systems is the diminutive Dell XS11-VX8, better known by its codename of “Fortuna.” If you are having trouble with the scale of this devices, that silvery part with the white label on the back of the server is the 3.5” hard disk. Yet this pocket-sized computer is no toy, but a true, enterprise-ready, 64-bit, hardware virtualization enabled, self-contained server with dual NICs and 2GB RAM that supports IPMI and iSCSI. The processor is the Via Nano, one of the most economical processors on the market with full hardware virtualization support for Hyper-V.

The post is a bit longer at http://blogs.msdn.com/the_power_of_software/archive/2009/06/08/little-and-large.aspx

Oracle and Google Netbooks Coming?

The Netbook market is one of the fastest growing IT segments.  All of these lightweight connected devices will push cloud computing use as these devices are developed for the always connected market segment.  The cel carriers are all scrambling to bundle 3G services with these devices as well.

The latest news is Larry Ellison discussing Netbooks coming from Oracle/Sun.

Ellison Mulls Foray Into Netbook Market

By BEN WORTHEN

Oracle Corp. Chief Executive Larry Ellison floated the idea that the software company might target mobile devices after its planned acquisition of Sun Microsystems Inc., including the small, low-priced computers called netbooks.

The possible focus on the netbook market was touched on in passing during a brief appearance by Mr. Ellison at Sun's annual JavaOne conference in San Francisco Tuesday. Netbooks are the fastest growing part of the personal computer business at the moment.

Associated Press

un Chairman Scott McNealy, left, and Oracle CEO Larry Ellison at the JavaOne conference in San Francisco on Tuesday. Mr. Ellison floated the idea of targeting netbook computers after Oracle buys Sun.

"I don't see why some of those devices shouldn't come from Sun-Oracle," Mr. Ellison said.

Acer announced Q3 availability of an Google Android laptop. Part of the differentiator is the energy efficiency from the ARM chip.

"Google really does have the brand name and the financial resources to be able to be a rival to Microsoft," said Warren East, chief executive officer of chip designer ARM Holdings. "The whole Linux community is a bit fragmented when you compare it with Microsoft."

ARM, based in Cambridge, England, is a potential beneficiary of Android's success because Microsoft's Windows XP, Vista and Windows 7 don't run on ARM-based computers.

"Microsoft going forward may have to work on an ARM-based solution," said Daiwa's Huang. "If Microsoft doesn't want to see Google Android get into the PC market, they will have to support ARM; otherwise, ARM will go perfectly with Android."

Qualcomm and Freescale Semiconductor, which make chips based on ARM's technology, said they expect to release products this year for netbooks that run Google's operating system.

Google's cachet will help get the computers into retailers such as Best Buy or France's Fnac electronics chain, said Henri Richard, chief sales and marketing officer for Freescale.

"It's important to be able to convince Fnac or Best Buy to put a product on their shelf, and that's where I think Android is going to be more helpful," Richard said in an interview.

SuperMicro 6016T-GF, Dual Nehalem and Nvidia GPUs, produce 2 Teraflop Performance

SuperMicro announced their Fastest 1U Server.

Supermicro Unveils Fastest 1U Server on the Planet at
Computex Taipei 2009

2-Teraflop SuperServer 6016T-GF Series with Dual Nehalem CPUs and
Two Non-Blocking PCI-E x16 Gen 2 slots for Two Double-Width GPUs

TAIPEI, Taiwan, June 1, 2009 - Super Micro Computer, Inc. (NASDAQ: SMCI), a leader in application-optimized, high performance server solutions, is introducing the fastest 1U server on the planet this week at Computex (booth N806, Nangang Upper Exhibition Hall). As the first 1U multi-GPU (graphics processing unit) system with a fully non-blocking architecture, Supermicro’s new 2-Teraflop SuperServer 6016T-GF Series features dual Nehalem processors and two Gen2 PCI-Express x16 interfaces to support two double-width GPUs.

"Our 6016T-GF SuperServers are by far the fastest 1U servers in the world," said Charles Liang, CEO and President of Supermicro. "Featuring advancements such as multiple x16 non-blocking native Gen2 PCI-Express connectivity, highly reliable thermal optimization, and industry-leading power efficiency, Supermicro has developed the world’s first truly optimized GPU-based servers."

It is interesting to think that the Nehalem chips are support chips for the GPU computing performance.

Who is using the Nviida cuda chip? pcmag has a bit of info.

Meanwhile, the new capabilities Nvidia's CUDA offers have attracted OEM partners. In November, for example, Dell, Penguin Computing, and Nvidiateamed up to launch a "personal supercomputer" that used the Tesla architecture that also drives the SuperServer 6016. Nvidia now has Tesla solutions for the data center, the departmental cluster, and for what the company calls the "scientific desktop" or the traditional workstation.

Meanwhile, over 1.2 million users now run CUDA-accelerated high-performance-computing applications, including oil and gas giants BNP-Paribas and Petrobas, as well as scientific research sites such as the Oak Ridge National Laboratory, Cambridge University, and the Max Planck Institute.

I discussed the idea of megacomputer in this post.