Intel Communities : All Content - Data Center Blog

Testing Live Migration with Intel® Virtualization Technology FlexMigration

webadmin@intel.com — Wed, 28 Jan 2009 19:13:39 GMT

Live migration is an essential technology for an agile, dynamic data center environment based on server virtualization. Until now, it has not been possible, however, to perform successful live migration between servers based on different generations of processors, each with different instruction sets - this limited our ability to implement large resource pools, creating islands of servers and hindering the implementation of advanced data center capabilities.

Combined, Intel VT FlexMigration assist and Enhanced VMotion are designed to overcome this limitation by enabling all servers to expose the same instruction set to applications, even if they are based on different processor generations from a single CPU supplier.

Intel IT and Intel’s Digital Enterprise Group, End User Platform Integration, conducted proof-of-concept (PoC) testing of live migration using Intel® Virtualization Technology FlexMigration (Intel® VT FlexMigration) assist and the Enhanced VMotion feature of VMware ESX 3.5U2*. All migrations completed without problems and our testing demonstrated that we can use Intel VT FlexMigration assist for live migration of Intel IT business applications in a mixed production environment. As a result, we can create resource pools that combine servers from multiple generations, eliminating incompatible islands of servers and allowing full implementation of advanced data center capabilities. Accordingly, we expect to standardize on systems with Intel VT FlexMigration assist in the future.

Our recently published IT@Intel white paper: Testing Live Migration with Intel® Virtualization Technology FlexMigration' documents the details pertaining to our tests – the types of systems tested, the workloads used, different scenarios examined and the results.

The paper can be downloaded at Testing Live Migration with Intel® Virtualization Technology FlexMigration

On behalf of our team, I’d like to invite you to view this whitepaper and comment about how you are using or intending to use these technologies in your datacenters and your experiences to-date with these capabilities.

Innovating the Thermal Management of High Density Servers

webadmin@intel.com — Mon, 22 Sep 2008 19:55:58 GMT

It's great server weather here in New Mexico today. The current temp is 74f , the high will be ~82f ,the low was 57f ,and the humidity will be ~30% all day. These are all "in range" of the air needed for a densely packed server to breath. Depending on how many servers are packed into a rack, they can heat this 74f degree air by as much as 50f degrees. Removing that heat from the server exhaust air consumes energy and expensive equipment. A key system trade off in data center design is that density reduces the unit cost of deploying and operating servers for most inputs but increases the cost of dealing with high density heat. If we can find ways to address this heat with less energy and cost then we can be more economical, and in today's words, more green.

It was on a day like today, a while back, where Tom Greenbaum, a fellow Intel engineer, and I were brainstorming how to get outside air directly to the servers. Because of Tom's experience in custom air handlers, we were focusing on economizers that are normally used for buildings. Economizers are commonly designed into office buildings, homes and in a simple fashion, most cars, but they are rarely are thought of for data centers. The key concept is that you have access to two air supplies: the outside air and the exhaust air. To be most economical, you want to continually decide "what is the best air to use next?". This takes measuring, deciding and switching. In your car, you are both the measurement device and decider as you press the re-circ button or the flow-through button on the dash. In a highend building air conditioner this is an outside weather monitor, an inside weather monitor, a simple controller and some extra duct work.

In our experiment we used an vane controller in a looped air duct that could blend outside air with the exhaust air to get us the best air for the servers to inhale at the lowest cost. The accuracy and controllability of our monitors and controllers were not as capable as we wished, but they did the job. The perfect controller would have measurements of outside air and exhaust air for: temperature, enthalpy and dew point. It would have policies that you could modify for best economy based on the requirements of your equipment. For example, it should be able to blend exhaust heat into the outside air to hold the minimum temperature at say 55f and no more in winter and it should be able to start incremental cooling loads as the temperature of the coolest of the two supplies rises above the maximum allowed by your equipment, in summer. It would decide to use some exhaust air when the dew point was higher than the outside temperature to control air humidity. You get the point, it would condition the air using all available inputs. Our PoC used DX cooling based units that usually are considered not as economical as water based cooling. But, in this mode of operation, they worked well and reduced complexity. In addition they used no water which is a plus in many desert locations. You can imagine evaporative systems in similar designs that could replace the DX units or work with the DX units for even more "economi-zation".

In the video you will see dust on the back of the servers. We had filtration on the air intakes and a control system that can indicate when the filter needs replacing, but we had a door system that let in dust near the inputs during really windy days. It was a design flaw in our temporary room. Once it got into the system we decided to let it run to see where it went. It collected in the exhaust areas but then created very little risk becasue most of the time we exhausted the air. Something that would not have been true in a closed loop system.

Don Atwood was able to negotiate for and create the production capable configuration for a sufficient number of production servers that were dense enough to run a proof of concept (PoC). These servers run high volume batch computing and are nearly always running above 90% utilized, perfect heaters for the job. It is important to note that this PoC was dependent on the concept of a "Compute Center". A Compute Center is the idea that high density servers can be isolated in their own air space a very short distance from the storage. The storage is left in the classical, and perhaps now more aptly named data center. Where this concept is able to be used, it can help free up traditional close loop environmental control for storage systems. If anyone knows of a great economizer controller, please lets us know. An Atom based design would be a plus

The temporary "compute center" we established and operated would not have be successful without the help of several great engineers contributing insight and innovation. Tom Greenbaum, Marvin Bailey, Steven Bornfield, Natasha Bothe, Greg Botts, Demetruis Ferguson, Ryan Henderson, Dan Links, Don Wright all contributed to learning what is possible during our PoC.

Well, it's 79f outside now and still great server weather, there are hot air balloons in the sky this morning, as we get ready for the Balloon Fiesta here in early October. Hot air balloons, like racks of servers, love this air becasue they can inhale cool air and then heat it to create work that moves people. Finally hot air balloons spend very little energy exhausting the resulting heat out the top. Its a simple model really, it just takes a smart controller.

Video: http://video.intel.com/?fr_story=2d6e0fbbef76b72c6119cc7fe7889bba20cb5192&rf=bm

Paper : http://www.intel.com/it/pdf/Reducing_Data_Center_Cost_with_an_Air_Economizer.pdf

Free Cooling for Data Centers - video and whitepaper

webadmin@intel.com — Fri, 19 Sep 2008 17:26:43 GMT

Hi I’m Don Atwood, author of the newly released white paper and video that discusses our proof of concept (PoC) that tested cooling our Data Center with outside air. The topic of humidity control and if this would work in an ultra high humid climate keeps coming up. Most OEM spec’s allow for a wide range of humidity and it’s our belief that this cooling methodology could be used almost everywhere globally. Our only uncertainty comes around trying this near the ocean with high levels of salty corrosive wet air. We know it would negatively affect the servers at some point but the question is how quickly and is it within our refresh timetable. During a trip to ASIA last week I discussed trying a small scale “near the ocean” PoC to test this theory.. Does anyone thing this would add value to your company?

Intel Lowers Energy Costs for High Performance Computing

webadmin@intel.com — Thu, 29 May 2008 23:15:28 GMT

The current uptake in high performance computing means mostly good things, but it also comes with a few built-in challenges. The paradox of this particular progress is this: when you scale hardware, you oftentimes scale power consumption, right along with it. My colleague, Shesha Krishnapura, a senior principal engineer from Intel IT Core Systems Engineering group, has some good news to share, in this podcast speaking with

 The Register’s ]]>

Tim Phillips. Shesha says that “In the past, that power relationship has existed. But with Intel’s core microarchitecture based platform, the power envelop has remained constant while performance has climbed significantly.”

Check out Shesha’s podcast below.

http://www.podtech.net/home/5166/intel-lowers-energy-costs-for-high-performance-computing

Intel Helps Developers with Multi-Threaded Software Community

webadmin@intel.com — Thu, 22 May 2008 19:42:53 GMT

In this podcast,

The Register's

Tim Phillips speaks with my colleague, Shesha Krishnapura, Senior Principal Engineer from Intel IT Core Systems Engineering group about developers' adoption of multi-core technology. Shesha sees that, while multi-core processors have become mainstream, now is an especially critical time for multi-threaded software, given the uptake in the industry and the increasingly urgent need for the software that will help to realize the higher performance potential of multi-core platforms.

Check out the podcast below.

http://www.podtech.net/home/5177/intel-helps-developers-with-multi-threaded-software-community

Before you throw it away....make sure you know where it is going.

webadmin@intel.com — Mon, 28 Apr 2008 16:27:39 GMT

I'd like to introduce myself--my name is Ilene Aginsky and I'm the new site community manager for the IT@Intel zone on Open Port, Intel's online IT community site. I started out in IT about two years ago and got very interested in the green aspect of IT.

We have had quite a few discussions out here in the community on green versus efficient Greening Data Centers or Make 'em Efficient? and I'm not as concerned by what it is called but rather what we must do to ensure we don't damage the environment.

The issues are not simple and require a balancing act. It is important to look at the picture holistically, from cradle to grave. For example, Intel IT will be refreshing approximately 20,000 servers this year with new servers that will consume less energy and reduce our carbon footprint. This begs the question: what happens to the old equipment and what are we doing to prevent it from ending up in a landfill?

I asked my colleague Robert who is the Secure Data Control Program Manager for IT and he told me that all end-of-life (EOL) servers at Intel follow the same process. We make sure that we secure all the data by removing and sanitizing the hard drives.

Once the data has been sanitized there are three possible paths:

Resale - we prioritize re-sale

Donations - some organizations need servers, even without disks

Scrap - anything deemed worthless to resale or donations is sent to scrap vendors for material reuse and recycling

What does your organization do with old equipment?

Intel IT Deploys Virtualization - How we did it!

webadmin@intel.com — Thu, 24 Apr 2008 17:42:12 GMT

As this is my first blog on this forum, I'd like to introduce myself. My name is Bill Sunderland and I have been working at Intel for 12 years primarily working on Server Hardware Engineering and the last three years of which I have focused my efforts on Program Managing the Virtualization Engineering release for Intel IT. I have recently published a WP demonstrating the methodology used as described below.

Intel IT planned, engineered, and has begun deploying a virtualized business-computing production environment at several data centers, a rollout that will continue through 2008. Our initiative has already confirmed anticipated virtualization benefits such as faster, more automated deployment. We are initially consolidating older servers running applications that are not mission-critical; we see opportunities to achieve 16:1 consolidation ratios.

Click here to read the WP: Implementing Virtualization in a Global Business-Computing Environment

I would be interested in hearing your experiences and/or questions regarding virtualizing IT environments!

Comparing Two- and Four-socket Platforms for Server Virtualization

webadmin@intel.com — Tue, 22 Apr 2008 23:19:40 GMT

The relative positioning of 2 and 4-socket servers for server virtualization has been an open question for a long time - a question that has stumped the most astute of IT professionals time and again. In fact it might not be an exaggeration to say that this open question is almost in the same class as the famous Riemann's hypothesis that has remained unsolved for over a century! (If you accept that premise, then there's some real estate on the moon that I'd like to bring to your attention as well). Although advocates for either class of servers have been emphatic in their respective positions, compelling data-points supporting their positions have been few and far between.

To remedy this sorry state of affairs, an Intel IT team conducted in-depth tests and analysis using current quad-core processor based 2 and 4-socket servers in a virtualized environment. This effort culminated in a comprehensive framework for comparing server platforms for virtualization. This comparative framework encompasses the majority of common deployment scenarios and usage models and answers - once and for all - the long unanswered question "which server is more appropriate for my virtualization project?"

The whitepaper detailing the findings can be found here Comparing Two- and Four-Socket Platforms for Server Virtualization. If time is short, click on the icon below for a short video overview.

The Three R's of the Data Center

webadmin@intel.com — Tue, 22 Apr 2008 15:11:27 GMT

My daughter recently brought home from school a photocopy of the lyrics of Jack Johnson’s “The 3 R’s” (from the Curious George soundtrack), which encourage us to “Reduce, Reuse, Recycle”. This struck me as relevant in some of the recent discussions I have been having about Greenwashing in the Data Center.

A fair amount of our data center strategy deals with driving down costs. We’re trying to spend less money to deliver the same or better results. Along the way, we find opportunities to be green. While I would love to have more meetings that start out with the question of “what can we do to help the environment?” rather than “what can we do to cut costs?”, we do talk about both. This is somewhat similar to consumer-oriented eco-efforts, encouraging people to save the world while saving money: unplug electronic devices when they're not being used, replace your appliances with more energy-efficient ones, etc. I don't know of many people or organizations who wouldn't like to spend less money, and when we can help the environment at the same time it's win-win.

Which brings me back to the song lyrics. Our cost-cutting measures tend to be related to at least two of the three “R’s” – reducing what we consume, many times by reusing what we already have. I’ll spend my next few posts exploring this a bit further, giving some specific examples of our cost-savings initiatives that ultimately contribute to a greener data center and IT infrastructure.

Happy Earth Day...

Part II: What if you invested a dollar and it returned 10?

webadmin@intel.com — Fri, 11 Apr 2008 16:41:25 GMT

Doug Garday continues his podcast series with part II, which continues the discussion around a heat recovery system to reduce the total cost of ownership. In this podcast, Doug plugs in numbers that show potential energy cost savings.

http://www.podtech.net/home/5079/roi-of-data-center-heat-recovery

To listen to Part I go to What if you invested a dollar and it returned 10?. View the full brief at Data Center Heat Recovery Helps Intel Create Green Facility.