Global Foundation Services - Data Center Team

Microsoft Cloud Fundamentals Video Series Available on Data Center Security

msdcblog — Fri, 25 May 2012 12:49:00 GMT

By Mark Estberg, Senior Director, Online Services Security & Compliance

Customers need to know that their cloud provider(s) are being responsible with the applications and data they entrust to them, while also being mindful of their requirements and whether they are compatible with the deployment model(s) they are evaluating. Read our blog and watch our new videos on this subject that is posted on our new blog site that is hosted on Microsoft’s Windows Azure platform! You can also find the latest insights, discussions about our cloud customer’s priorities, news and more from the Microsoft Global Foundation Services team on our new web site here.

Please don't forget to re-subscribe to the RSS feed on our new blog site for all the latest news and posts.

//me

Visit our New Blog and Web Site now hosted on Windows Azure!

msdcblog — Thu, 17 May 2012 22:00:00 GMT

We are pleased to announce our web site and blog are now located on one site that is hosted on our Windows Azure platform! You can find the latest insights, customer priorities and news about Global Foundation Services.

Don't forget to re-subscribe to the RSS feed on our new blog site for all the latest news and posts. More about the site is outined below.

The Site's content includes:

Customer Priorities Pages that dive deep into infrastructure architecture, operations, reliability, security and compliance, and the sustainability strategies for our data centers and fiber optic networks – all top of mind for customers, including:

20-30 minute interactive presentations featuring our data center subject matter experts addressing some of the top 3-5 customer questions we receive for each key customer priority
Strategy Brief documents addressing each priority and presenting the best practices of our engineers -- for downloading and sharing
Video overviews from some of the company's strategy leads addressing the key customer priorities
Related whitepapers, technical content, and slideshows
Video tours of our data centers

A Future Insights Page that explores the forward-looking challenges of providing cloud services at massive scale and the business impact, reliability, security, sustainability opportunities and roadmap discussions

Our Team Blogs with industry relevant discussions on cloud infrastructure issues and best practices. We have migrated all our blog archives to our Azure-hosted site and will be blogging regularly there now.

We’ve been sharing our best practices with the industry since 2007, and now we have re-launched our blog and web site with these additional resources to help us scale to meet the needs of our customers and the industry who want to better understand Microsoft’s data center strategy. These new resource also help us to more broadly share our best practices as we are rapidly evolving and making strides towards more operationally robust infrastructures that are efficient, sustainable, reliable, and secure.

We hope you will enjoy the new site and resources, and share them with others!

Making Carbon Neutrality Everyone’s Responsibility at Microsoft

msdcblog — Wed, 09 May 2012 23:54:00 GMT

Yesterday, Microsoft announced a significant step to reduce our environmental footprint by committing to achieve carbon neutrality by 2012. As the team that operates our cloud infrastructure, Global Foundation Services (GFS) will be instrumental in achieving this goal.

First, our continued effort to improve efficiency in our data center operations, using such measures as Power Utilization Effectiveness (PUE), and Carbon Utilization Effectiveness (CUE), will help reduce our overall carbon footprint. Second, our detailed measurement and chargeback model allocates expense to business workloads based on energy consumption, establishing the right incentive to design for efficiency. We are proud to be taking these steps and driving accountability for reducing our environmental impact across all of our business units.

We believe climate change is a serious challenge requiring a comprehensive and global response from all sectors of society. For more information on our carbon neutral strategy, read our whitepaper, “Becoming Carbon Neutral: How Microsoft is Becoming Lean, Green, and Accountable.” You can also read more about this and our other environmental efforts on Microsoft’s blog Software Enabled Earth.

(Don't forget to re-subscribe to the RSS feed on our new blog site for all the latest news and posts.)

//cb

Microsoft Recognized by EPA as Leading Green Power Purchaser

msdcblog — Sat, 28 Apr 2012 02:27:00 GMT

This week, the U.S. Environmental Protection Agency (EPA) released a list of the Top 50 Green Power Partnership organizations, listing Microsoft as the third largest purchaser of green power. The EPA evaluates organizations voluntarily using clean, renewable electricity from resources such as solar, wind, and low-impact hydropower. As Robert Bernard, Microsoft’s Chief Environmental Strategist, details in his environmental sustainability blog post, ‘Microsoft earned this recognition by purchasing more than 1.1 billion kilowatt-hours (kWh) of green power annually. This is enough green power to meet 46 percent of Microsoft’s electricity use and is the annual equivalent to taking more than 150,000 passenger vehicles off the road’.

As Microsoft continues to invest in and explore opportunities for renewable power and greater efficiencies across our infrastructure, our data center will play a large role. As I noted in my recent blog post, the
necessity for clean, renewable, reliable, and low cost power will increase exponentially in the near future. This has led us to look at how we can power the cloud infrastructure off the grid through the integration
of power plants and data centers.

As we continue to make strides towards the Disappearance of the Data Center and reducing our carbon footprint, we are pleased to be honored by the EPA for our work in this area and look forward to reporting on future advances.

(Don't forget to re-subscribe to the RSS feed on our new blog site for all the latest news and posts.)

//cb

Thinking Off the Grid: Independence for today’s Data Centers via Data Plants?

msdcblog — Wed, 18 Apr 2012 14:00:00 GMT

By Christian Belady, General Manager of Data Center Services

Question. Why do data centers need to be connected to a dirty, expensive, unreliable electrical grid? Answer. They don’t and they don’t want to be either. Integrating a data center directly into the power plant -- what we are calling our Data Plant program -- will allow a data center to pick its sustainable fuel source and shield itself from grid volatility.

As customer demand for more and more cloud services continues to grow, Microsoft, as well as many other IT companies, will need to aggressively plan for how we can significantly reduce our carbon footprints and make our data center operations more efficient. The necessity for clean, renewable, reliable, and low cost power will increase exponentially in the near future. Without a bold shift in strategy, our entire industry will become more dependent on a costly, antiquated, and constricted power grid.

The future requires a fundamental shift in how we approach data center design and how we source energy for our data centers. I still think the biggest opportunities are ahead of us. Today, Microsoft is focusing on integrations that will drive down costs and improve sustainability long term:

Integration from the Chip to the Utility
Integration across the Hardware and Software stack

Our team has been focusing on this since last year, and right now it's really about how do we integrate even beyond what we've done already? How do we integrate from the chip all the way out to the utility, but not only from a hardware perspective? How do we integrate across that whole stack and make the right tradeoffs to drive the most sustainable, affordable solution?

So what is a Data Plant? In 2010, I started introducing the idea of a Data Plant at industry speaking engagements, as a grid independent data center and the integration of power plants and data centers. This concept opens up a host of options for a data center operations team, allowing it to target clean, sustainable, cost effective, and reliable fuel. A Data Plant can be integrated with the grid for back-up, or in an island mode to facilitate locations that are not near large transmission lines. As an option, a connection to the grid would allow the power plant to maintain a maximum capacity factor; in other words any unused power can be sent back into grid, keeping the power plant fully utilized at all times and eliminating the risk of stranding capacity.

Currently, our team is researching the first-ever grid independent fuel cell, data center that is fueled directly from biogas. The experiment is small scale, so we can demonstrate and measure the benefits of it like we did with our “data center in a tent project” in 2008. We are also talking with several municipalities about a public-private partnership to test a prototype. The impact of this Cleantech project is significant. For one, biogas is 100 % renewable. A small 200 kW prototype data center will offset over two million pounds of CO2 emissions per year which is the equivalent of about 300 Honda Civics being taken off the road. One can start to see how significant this program will be when it’s implemented in the megawatt scale.

Where do you find biogas? Landfills and water treatment plants are two of the most common sources, although there are many others. Water treatment plants are mission critical installations that produce methane (CH4), a greenhouse gas that is 20 times more potent than CO2, as the sewage from our communities are broken down in an anaerobic digestion process (decomposition without oxygen). Landfills produce methane in a similar way as our garbage slowly decomposes underground. The methane that is produced by both approaches must be flared, converting it to CO2 to minimize the impact on the environment.

But haven’t I heard of other data centers powered by biogas? This data center will be constructed directly on a water treatment plant or landfill site and will consume waste gas that is typically flared off. Some data centers are reported to be fueled with ‘Directed Biogas’ where biogas is injected into the natural gas pipeline and then natural gas is purchased at the data center site many miles away. The Data Plant consumes the biogas as it is produced onsite making it purely biogas powered.

Why Fuel Cell? The process used in fuel cell technology yields a very reliable source of energy while producing a fraction of the emissions that other generation technology emits. The clean electrochemical reaction of this device makes it a much cleaner and easier to secure air quality permits because there are no carcinogenic DPM (Diesel Particulate Matter) to worry about. Compared to internal combustion type generation technology, a fuel cell’s Nitrous Oxide (NOx) and Sulfuric Acid (SOx) emissions are barely measurable. Fuel cells basically exhale the same thing you and I do, CO2 and water vapor.

In addition to the clean technology benefits, fuel cells offer us a host of other advantages including increased Mean Time Between Failure (MTBF), higher availability, pay as we grow, flexibility in fuel sources, high electrical efficiency, combined heat and power source, and reduced dependence on traditional data center infrastructures such as five minute UPS and back-up diesel generators.

Simplified Design. Ironically, the unreliability of the grid requires us to install a complex array of UPS, back-up generators, maintenance bypass circuits, power conditioning, etc., that adds additional impending sources of failure. Keep in mind, a data center does not actually have to incur a utility outage for there to be problems. Spikes and sags at the millisecond level can result in component damage downstream. The reality is data centers are constantly bombarded with power quality events and transients that over time degrade the built in protection infrastructure of the data center lowering confidence in older, well-engineered data center electrical plants.

As the demand for cloud services grows, we are looking at new methods for maintaining the high availability of our applications, while becoming more sustainable, efficient and cost effective so we can pass those benefits on to our customers and our shared environment. A constraint we all need to work with, is the fact that our electrical grid was never methodically planned or engineered for the significant growth we are experiencing today. And it certainly was not engineered to take on the proliferation of data center growth. Independence from the power grid will allow our industry to minimize its impact and ease some of the constriction already taking place. The Data Plant is one way of giving us an ability to manage the growth of our clouds in a thoughtful manner: building in sustainability from the ground-up, so we can run sustainably every day. Our goal is to reduce the impact of our operations and products, and to be a leader in environmental responsibility.

So, in parting let me give you my final thoughts. Clearly, the industry is going to face some challenges with respect to power, carbon, and water as a resource. Within Microsoft, we are working to proactively address these issues. You can expect massive integration in cloud infrastructure, and it's purely driven by sustainability and a total cost of ownership advantage. Remember, lower costs generally mean better sustainability too. If you drive your costs down, you are actually using less material and energy. That also makes it more sustainable. Think about that in terms of the data center infrastructure and about data as the next form of energy. How would you rethink your operations if you did that? As I've said before, there's tons of opportunity and there's certainly a lot for all of us to learn. This industry is blessed with a lot of smart people attacking the same problem set. Personally, I am very excited about these possibilities and believe I am fortunate to be working on this technology at this time.

Our team will continue our efforts through R&D, blogs, involvement in industry organizations, speaking at conferences, and white papers to share what we are doing in the most transparent way possible. We have been busy in the past year creating all the content you will find on our web site today. I hope you will explore the videos and arsenal of related papers from our engineers and subject matter experts published on it. I hope these will continue to stimulate a healthy discussion and continued collaboration with our industry participants.

(Don't forget to re-subscribe to the RSS feed on our new blog site for all the latest news and posts.)

//cb

Operational Excellence is Essential for a Reliable Cloud Infrastructure

msdcblog — Mon, 16 Apr 2012 17:47:00 GMT

Attributed to: Christian Belady, General Manager of Data Center Services & Vijay Gill, Senior Director of Network Engineering

When Microsoft delivered its first online service back in 1994, the cloud was still a relatively undefined, evolving space. Today, Microsoft's cloud supports more than 1 billion customers and 20 million businesses in 76 countries and it's still growing. To support the increasing array of enterprise-level online services, Microsoft is investing heavily in a global cloud infrastructure that delivers security, reliability, scalability, efficiency, and performance. A key concern among customers is that their services are persistently available, and that their cloud provider has the technology, policies, and processes in place to maximize uptime.

Disruptive events such as natural disasters, power grid outages, and human error are a fact of life, so it is critical that our cloud infrastructure is architected for resiliency, our processes result in rapid problem resolution with minimal customer impact, and we maintain a culture of service diligence and continuous innovation to identify and eliminate systemic weaknesses.

Laser-sharp focus on operational excellence helps Microsoft ensure reliability, availability, and security 24x7x365. Operational excellence can be complicated, and our customers often want to know what it encompasses.

An effective cloud infrastructure strategy pivots off of two things: keeping our sites up, while driving costs down. We architect a resilient cloud infrastructure with geo-redundant back-up for recoverability and data integrity. The company is making a heavy annual investment in over $9 billion in research and development on our cloud, services, and software to ensure a strong innovation roadmap.

The Microsoft cloud is also founded on robust security policies and procedures that differentiate data (consumer vs. enterprise), separating private company information from commercial use. We maintain a fully developed Service Management framework called the Microsoft Operations Framework which guides and formalizes our compliance with international standards such as ISO20000 for Service Management, ISO 27001 for Information Security, and BS 25999 for Business Continuity. In addition, we hold third party, process-based attestations and certifications including SOC (SSAE 16/ISAE 3402), FISMA, Sarbanes-Oxley, and HIIPA/HITECH.

Our continuous growth reflects judicious data center designs that add capacity and reduce attendant costs to maximize compute capacity with the fewest servers while maintaining workload isolation. We partner with leading hardware manufacturers to right-size the thousands of servers, routers, and equipment within them to balance the degree of physical device fault tolerance with the application fault tolerance.

We find the right balance of hardware and software resiliency to provide the best availability for the lowest possible cost. For Example, our network-one of the largest and most well-connected in the world-includes multiple, physically diverse connections into our data centers, and we maintain sufficient capacity to handle large scale network interruptions without degradation of performance.

Sophisticated service instrumentation and monitoring integrates at the deepest levels with each component, giving us visibility into the data center, network backbone, internet exchanges and beyond, to help us spot, diagnose and manage the cause of any disruption and resolve it quickly.

Technical troubleshooting expertise from our globally distributed Microsoft Operations Center's team provides round the clock staffing, failover capabilities and the resources needed to triage, mitigate and escalate issues as they unfold in real time. The way that we manage releases and changes through formalized processes helps to remove the potential for human error, leverages standardization and protects data in a vigilant manner.

One of the world's largest fiber backbones powers our data centers, providing more than 3.5 terabits per second of capacity to more than 1200 networks with robust availability. It provides the ability to instantaneously reroute around internet failures and the capacity to withstand significant network interruptions.

Finally, consideration of environmental issues and a relentless focus on energy efficiency is a final component of Operational Excellence. The Microsoft data center operations team continually seeks to drive more efficient power use and cooling in our data centers. Our latest modular data centers use about 50 % less energy than those from three years ago. Our data center designs - pioneering the use of fresh air cooling and ultra-efficient water use in the latest modular facilities, use only 1 % of the water used by traditional data centers in the industry. They also use recyclable materials in construction and build in sustainability measures that ensure we are constantly gaining in efficiency and sustainability. (In fact, we were honored that the United Nations Environment Programme relied on our modular data center designs for the IT infrastructure at their new carbon neutral headquarters complex.)

We also drive efficiency and sustainability in how we operate our data centers. Microsoft's are among the most monitored and measured data centers in the world, informing more efficient operations and identifying areas for future research. At the request of the US Environmental Protection Agency, we shared how we do this at their industry stakeholders meeting on efficiency as an example of how to do it right. Also, by eliminating unnecessary components, using higher efficiency power supplies and voltage converters, and binding the expandability of server platforms, we achieve significant power savings. We look at specific measures such as processor performance per-dollar, per-watt to determine optimum tradeoffs in processor selection. We also operate our servers at a wider temperature range and use free air cooling and water economization to improve efficiencies. In San Antonio we have been using recycled waste water since 2008 to cool the facility.

With all these opportunities for efficiency, we were not surprised that when we provided data to Accenture and WSP Environmental for a study on the environmental impact of cloud computing, they found that when organizations move their Microsoft business applications to our cloud, there's a net energy savings per user of at least 30 %. For small businesses the result can be even more dramatic, with potential carbon savings of up to 90 % per user.

As part of our commitment to driving not only greater efficiency, but also reducing the impact of our data centers for the long run, later this week we will be sharing some insights into an exciting alternative- energy research project we have been working on. While we have, and continue, to purchase green energy at many of our locations around the world, we have a fundamental belief that there is a potential to redefine how we and the industry power our data centers.

A few years ago, we spoke about the concept of looking at data as a form of energy and the concept of data plants. Last July in a blog called The Disappearing Data Center we expanded the concept further. Through massive integration of power plants and data centers, we felt that there could be huge efficiency gains by eliminating the need for transmission lines, substations, and transformers (as well as the associated transmission losses) that we see in today's power distribution ecosystem. With these data plants we distribute data (an energy form) in a network (an optical grid) providing the next generation of energy distribution. Looking at it this way, we are essentially taking another step in the evolution of refining the energy being distributed. The point is that if we look at data as a form of energy, how does that change what we are doing today? When we talk about the disappearing data center, what we really mean is that it will disappear as we know it through integration and drive unprecedented levels of efficiency gain…. and this can only be done at the scale of the cloud. The figure below shows this evolution taking sustainability to new levels with other side benefits in terms of reliability and the ability of using waste gases.

The evolution of data into energy.

We are committed to the long term sustainability of our industry and believe this is where the industry should focus rather than short term manipulation of carbon PPAs. We have already invested millions in research in this area and have filed many patents in this area, including one on the Data Plant. We will provide more details later this week on these innovations via our team blog here.

As we continue to experiment and share our findings, we hope that we can accelerate ours, and our industry's continued move towards increasing positive computational impact while reducing operational impact.

We have come a long way since we built our first data center in 1989 (on our Redmond, WA campus) and launched our first cloud service MSN in 1994. We continue to make positive strides toward even greater reliability, availability, security, and environmental sustainability every day. Our team is passionately committed to not only improving our own cloud infrastructure designs and the operational practices of our facilities, but also sharing those with other stakeholders in the data center industry.

//cb and vg

Microsoft’s Dublin Data Center Grows with Enhanced Efficiency and Sustainability

msdcblog — Mon, 05 Mar 2012 15:00:00 GMT

By Dileep Bhandarkar, Ph. D, Distinguished Engineer
Global Foundation Services, Microsoft

Last week, we announced an addition of 112,000 square feet and 13.2 MW of cloud computing power to our Dublin data center. This facility and one in Amsterdam are some of several that we have that support the EMEA region. This current expansion announcement raised some questions about whether the new design is a modular or traditional build. How is it different from other modular facilities that we have built in other locations? Why didn’t we use ITPACs as we did in Quincy, WA? I hope this blog will help answer these questions.

The new modular Dublin data center being built next to the first facility continues our commitment to reducing material and energy consumption.

In July 2011, I wrote:

“In recent discussions at several data center conferences, there is still some confusion about what modularity means. To some it means containers. The dictionary defines it as “the use of individually distinct functional units” or “constructed with standardized units or dimensions for flexibility”. A modular data center consists of smaller building blocks that can be replicated to deliver additional capacity. The form factor and size (power and floor space) can vary greatly. Our Gen 4 data center designs feature modular buildings that function as “colocation rooms” that can be populated with servers as needed, while the ITPACs are shipped fully populated with servers. Both are free air cooled in selected facility locations. This is the next step in our pursuit of higher energy efficiency. The physical form factor and technologies will evolve over time.”

Let’s start with a brief background about our Dublin data center. The first phase came online in September 2009. It was our first mega data center outside the United States and we wanted to take advantage of the cool weather conditions that enabled a more sustainable outside air cooling approach.

Microsoft’s mega Dublin data center went live in 2009 and has won numerous Sustainability and Innovation awards (listed below).

The roof top air handling units (AHUs) push the outside air into server rooms that do not have raised floors. The PODs in the colocation rooms isolate the hot aisle and then the air handlers pull the hot air up to the roof and push it out of the building. When the temperature is cold, we mix warm return air with the outside air to maintain a 75° F temperature in the server rooms. Dublin was our first attempt at airside economization and there was some concern about our ability to operate using free air cooling exclusively. Hence, we added evaporative washers in the air handling units (AHUs) for supplementary adiabatic cooling and we added direct expansion (DX) units to our design to provide additional back up as needed.

Colocation PODs contain the hot aisles of servers.

During the first year of operation, the weather never required us to use the DX units. Reviewing the historical weather data showed that the warmest average high temperature in Dublin is 66 °F in July and August, and hasn’t exceeded a maximum of 82°F between 1971 and 2000. When we began planning for the capacity expansions, we determined the DX units were unnecessary and that the adiabatic cooling would provide ample backup when temperatures exceed 75° F. For point of reference, the standard maximum recommended temperature for data centers under guidelines from ASHRAE is 80.6° F.

We had several goals for the capacity extension currently under construction. We wanted to continue to meet the growing demand for our online services such as Bing, Office 365, MSN, Windows Live, X-Box Live and the Windows Azure platform, while advancing our energy efficiency and environmental sustainability strategies. We decided to build a single story, steel-frame structure that is sheeted with a pre-manufactured, recycled metal cladding. A concrete pad foundation and floor support the structure. A concrete roof will hold the roof top AHUs. This is a departure from the more traditional two-story construction used in the existing facility that has been operational since 2009. As a result of the modular design features, we are able to significantly reduce the materials used, carbon footprint and wastes by approximately 50% percent of that used in traditional builds.

To further improve the energy efficiency, we eliminated the DX units and the adiabatic cooling and return air fans from the new air handling units. As a result, the data center does not require or use any water for cooling. This reflects a growing body of evidence that data center facilities can operate at significantly higher temperatures (ASHRAE now defines an Allowable A2 range of up to 95° F) without damaging the IT equipment. By raising the operating temperature of facilities, significant savings in energy consumption can be achieved. We plan to operate both Dublin facilities at 75° F which is well within the recommended range set by ASHRAE – with infrequent excursions during short hot spells to 85° F. Most server manufacturer’s warranty their equipment up to 95° F.

We are also using the free air cooling approach pioneered in our Dublin facility at newer modular sites in Boydton, VA and Des Moines, IA. However, the Dublin facility does not use the ITPACs that are being deployed in our Quincy, WA data center for specific workloads. The ITPACs use a different form factor, but the outside air-based cooling approach is very similar.

Our strategic modular design approaches are rapidly evolving as demand for our over 200 global cloud services grows. Therefore, our deployment of modular form factors in our infrastructures is broader than just ITPACs. Not many cloud providers today deploy and manage a cloud services portfolio as huge as ours and we have learned a lot about running data centers since 1989 and hosting cloud services since 1994. For operating at this huge scale, it’s about mechanizing the construction process of the entire facility and building for scale, reliability and availability for the services you are providing. The Dublin facility is designed to meet the needs of a wide set of workloads for the EMEA region and the modular form factors are selected to support specific services and SLAs hosted within it.

ITPACs were deployed to support cloud services at the Quincy, Washington data center.

The design also enables future growth in capacity. The facility can be fitted and brought online in 4.4 MW increments, making it computing scale modular as well. Given the wide range of online services and anticipated growth, this was the right decision for our customers.
The data center has gained the following recognition from the industry for its sustainability and innovation design strategies:

Best European Enterprise Data Centre Facility by Data Centres Europe 2010
Data Centre Leaders Award for Innovation
European Commission’s Code of Conduct for Data Centre Sustainability Best Practice

Our team is proud of the improvements made and efficiencies gained. We look forward to bringing the new 112,000 square foot facility online around the end of 2012.

//db

Computerworld names Microsoft one of Top Green IT organizations

msdcblog — Mon, 24 Oct 2011 21:09:00 GMT

Cross posted from the Software Enabled Earth blog

Today IDG’s Computerworld named Microsoft as one of the top green IT organization as part of its annual Top Green-IT Organizations feature. Of the seventy organizations that participated in the ranking, Microsoft was ranked fourth. We’re very pleased to rank among other leading organization committed to environmental sustainability and green IT practices.

Computerworld’s criteria for top green IT companies looks at how organizations are reducing energy consumption in IT equipment and how they’re using technology to conserve energy and lower carbon emissions. Other factors, such as corporate practices including recycling programs and telecommuting, are considered as well.

Microsoft has a current goal to reduce carbon emissions by 30% per unit of revenue by 2012 based on 2007 levels. Our current carbon reduction efforts are focused on three primary areas: facilities, travel and data centers. Some of our recent successes include:

Efficient Buildings: All new Microsoft-owned buildings are designed to Silver or greater LEED standards and consume 20% less energy than traditional buildings. Microsoft is currently in the pilot phase of a “smarter buildings” program that uses IT to better manage energy efficiency across our campus. Early results have shown that an investment of 10% of annual energy expenditure will pay itself back in less than 18 months. We’ve also launched a “Sustainability Champions” program, an initiative for Microsoft employees to rally their co-workers to adopt more sustainable work habits. The goal is to reduce energy consumption in all Puget Sound buildings by 3 to 5 percent.
Waste Reduction: Over the years, Microsoft has replaced polystyrene and plastic flatware with compostable products in offices throughout the Puget Sound, thereby reducing our on-campus waste by 50%. We’ve also significantly increased our diversion away from the landfill for waste. We are currently at 61% diversion in the Puget Sound and are targeting a goal of a 70% diversion rate for waste by 2012.
Travel: Microsoft’s private bus service—which provides employee commute service to our main campus in Redmond—has reduced Microsoft’s single occupant vehicle rate from 66% to 62% in one year and saved 3,000 metric tons in carbon emissions. We’ve also adopted a new international company car policy that requires CO2 emissions for all new Microsoft company cars to average 130 grams per kilometer, representing a significant reduction from previous averages. On our Puget Sound campus, we’ve recently added 12 electric vehicle charging stations to the campus to support employee-owned electric vehicles, as well as other company service vehicles.
Data Center & IT Efficiency: Microsoft has emerged as a leader in data center operations. We recently opened a new “Generation 4” data center in Quincy, Washington, which is achieving a PUE (power usage effectiveness) of 1.15 to 1.20 and gets its electricity from hydropower. We are on track to achieve our goal of designing and constructing data centers that average 1.125 PUE by calendar year 2012. We’ve also stepped up efficiency efforts for PC’s/laptops that employees use. We’ve reduced energy use by 32% in PC/laptop power consumption and our internal PC standards are 100% Energy Star compliant.

In addition, our cloud computing services enable other companies to take advantage of Microsoft’s ultra-efficient data centers. In a recent study we conducted on the environmental impact of moving Microsoft Exchange, Microsoft SharePoint, and Microsoft Dynamics CRM to the cloud revealed that customers who choose to run these applications in Microsoft’s cloud versus on-premises can reduce energy consumption and carbon emissions on a per-user basis by at least 30 percent versus running those same applications on-premise. The benefits are even more impressive for smaller businesses who have the potential to realize energy and carbon savings of more than 90 percent.

We are certainly proud of the work we’ve accomplished and the progress we’re seeing at our company. We are flattered that Computerworld named Microsoft as one of the top green IT organizations, but we acknowledge our critics too. Do we have more work to do? Absolutely. We see the opportunity to improve in a variety of areas and remain committed to doing so. We plan to continue making the right investments to ensure that Microsoft remains an industry leader in green IT.

Microsoft’s Data Center Takes Fresh Approach On Water Reuse

msdcblog — Thu, 13 Oct 2011 17:00:00 GMT

Today we are transferring our $ multi-million water treatment plant to city of Quincy, WA

By: Christian Belady, General Manager of Data Center Advanced Development

Around the globe, water is becoming a scarcer and more valuable commodity, and that’s an important factor for data center operators and cloud service providers to consider as consumers and businesses aggressively adopt cloud-based computing. It’s even more critical that all of us in the industry make sure that beyond building sustainability into our designs, running data centers to higher standardize efficiencies, and measuring impact constantly, that we are helping the industry at large in thinking out of the box.

Today offers one of those opportunities. In Quincy, Washington, we are taking steps to transfer the operations of our Water Treatment Plant, located on our data center site, to the City of Quincy. This project involves innovative agreements for promoting a long term sustainable use of a limited natural resource, water, in a desert area that has the added benefit of supporting the foundation of Quincy and Grant County’s growing economy for years to come. To my knowledge, it is the first known transfer of a water treatment plant to a municipality in our industry and I would like to share why I think this type of collaborative project helps the industry and environment benefit as a whole.

Microsoft’s Quincy, Washington Water Treatment Plant

Microsoft’s current water treatment plant extracts the minerals from the potable water supply prior to using that water in cooling our local 500,000 square foot data center. (Note: our new modular Quincy data center that went live in January 2011 uses airside economization for cooling and substantially less water).

Moving forward, the City of Quincy will lease the water treatment plant from Microsoft for $10 annually and will provide the company with reduced water rates, with an option to buy it after 30 years. The plant will be operated, maintained and managed by the city with a right to purchase the plant after that time. By loaning these assets to the City, they were able to save significant construction costs for the new Reuse System.

The strategic location of the water treatment plant will also benefit other local businesses and industrial users, such as other data centers, food storage processing companies, etc. The City of Quincy plans to retrofit the plant as an expanded industrial reuse system in two phases. Following the first phase, the system will generate approximately 400,000 gallons per day (150 million gallons per year) using food processor wastewater effluent. The second phase upgrade is projected to produce 2.5 to 3.0 million gallons per day (1 billion gallons per year), with about 20 percent being used by local industries and the remaining being used to recharge the aquifer around Quincy.

This collaborative partnership with the City of Quincy solves a local sustainability need by taking a fresh look at integrating our available resources and allows Microsoft to focus on its core expertise in meeting the needs of its customers who use our Online, Live and Cloud Services such as Hotmail, Bing, BPOS, Office 365, Windows Live, Xbox Live and the Windows Azure platform.

As I have said before, we will continue to look for ways to reduce and eventually eliminate the use of resources, including water, in our data center designs. Today, this project reflects our firm commitment to that vision. I hope it also sets the stage for a healthy discussion within our industry to continue to explore ways to share our investments and best practices within the industry and regions in which we do business.

For more information on our Quincy, WA data center please watch our data center tour video.

//cb

Microsoft’s Trustworthy Computing Blog Site Launches Today

msdcblog — Fri, 16 Sep 2011 04:08:17 GMT

Microsoft’s Trustworthy Computing team just launched their new blog site, where you can find an on-going discussion on trustworthy cloud computing at Microsoft and throughout the industry.

The goal of this new blog is to give readers a look into how Microsoft works to build and maintain security, privacy and reliability in everything the company does. The blog will often touch on these subjects with a focus on Microsoft’s cloud operations.

It’s an open dialogue community site, so please feel free to suggest posts or ask questions. You might even catch a guest post from our team in the future, so keep an eye out!

//GFS Data Center Services Team

Microsoft Shares Video Tour of its Cloud Datacenters

msdcblog — Tue, 26 Jul 2011 06:59:00 GMT

By Dileep Bhandarkar, Ph. D.
Distinguished Engineer
Global Foundation Services, Microsoft

Today our team is releasing a video tour of four of Microsoft’s global datacenters. The video provides an inside look at how the company has invested heavily to rapidly evolve its datacenter IT infrastructure and best practices to deliver greater scalability, reliability, efficiency, security, and sustainability.

We have come a long way since our first datacenter in 1989. The server count, power efficiencies and number of facility locations have grown dramatically. Microsoft’s cloud is now comprised of a globally distributed datacenter infrastructure supporting hundreds of online services, including Windows Azure, Office 365, Bing, MSN, Windows Live, Hotmail and Xbox Live. More than a billion customers and 20 million businesses in over 70 countries use these services each year.

To meet our business goals for delivering higher energy efficiency, sustainability, and computing capacity at scale, we started designing our own datacenters and servers a few years ago and have been constantly innovating to improve upon our best practices and key learnings gained over the past 22 years. Our first Microsoft-designed datacenter opened in 2007 and it gave us better control over our operating environment. This allowed us to achieve Power Usage Effectiveness (PUE) well below the then industry average of over 2.0 for traditional facilities.

As our user base grew, operational excellence and services reliability required high availability through advanced monitoring and incident response, service support, and back-up failover capability, managed through our geographically distributed Microsoft Operations Centers—operating 24x7x365. Our customers demand trustworthy cloud services—requiring a comprehensive approach to security ranging from software development and operations to physical security. In addition to maintaining high standards for operations processes, we routinely bring in third parties to verify our capabilities, resulting in ISO and PCI certifications, SAS70 Type II attestations, and FISMA certification and accreditation.

Our pursuit of energy efficiency and cost effectiveness has been an ongoing activity. We divided our Chicago datacenter into two distinct level of service. The bottom floor houses water cooled containers without redundancy in the infrastructure. It relies on software that is resilient to failures. The container- based deployments allow us to deploy up to 2400 servers quickly, while driving the infrastructure cost down compared to traditional “N+1” datacenters. Our container bay is operating at a PUE of less than 1.25 compared to the traditional co-locations upstairs that operate at about 1.5.

We went on to build our Dublin datacenter with free air cooling to achieve even higher efficiencies for a traditional “N+1” datacenter with PUE of 1.25 compared to 1.5 for the upper floor in Chicago.

In 2010, we began building our first Generation 4 datacenter, culminating over three years of research and development internally and externally with industry partners. Generation 4 designs also built upon our Dublin datacenter experience of using free airside economization and applying it in a truly modular manner using two different form factors. The original container deployments have evolved into free air cooled Information Technology Pre-Assembled Components (ITPACs) that drive these new datacenters’ PUEs even lower than the Chicago facility containers. We are also using free air cooling for our N+1 modular designs, but now with simple metal buildings without raised floors. These fully modular facilities have reduced our time to market by 50 percent and deliver outstanding PUEs, low water utilization and low total cost of ownership.

In our Quincy datacenter, ITPACs are moved into place and connected to each other for complete module deployment.

The team is excited about the work ahead to support our growing cloud computing services. Our datacenters will continue to evolve to provide a robust, secure, highly reliable, and efficient foundational infrastructure. And we hope that by sharing our key learnings and best practices that we can help the industry, as a whole, to drive greater efficiencies in their environments and through the use of our cloud services.

We hope that you will enjoy the video tour of four our cloud infrastructure facilities.

\db

You can access more of our best practices around datacenter efficiencies, rightsizing servers for production environments, cloud security, etc. in published papers and blogs available on our web site at http://www.globalfoundationservices.com.

The Cloud Infrastructure Journey - From Containers to Modularity, Long Live Containment!

msdcblog — Wed, 20 Jul 2011 22:29:00 GMT

By Dileep Bhandarkar, Ph.D.
Distinguished Engineer
Global Foundation Services, Microsoft

In late 2007 we were intrigued by the opportunity of deploying servers in ISO (shipping) containers. Our analysis showed that we could pack as many as 2500 servers in a single 40 foot container and reduce our deployment times and improve efficiency. We started creating space for stacked ISO containers in our Chicago datacenter and developed a functional specification for the containers that allowed our OEMs to innovate while still meeting our physical and functional requirements. Our container infrastructure specification defined the physical space (length, width, and height), weight, power budget and voltage, inlet water flow rate and temperature, and return water temperature. Beyond this the OEMs were free to design the internals of the containers as they preferred, but they had to avoid stranding power. We specified the server configurations and network architecture. Our first deployment in 2009 resulted in up to2500 servers to be pre-installed and networked in a 40 foot ISO container. The containers were lifted off a truck and rolled in on air skids into their parking spots and connected to power, network and water in a matter of a few hours.

Moving a container into the Chicago datacenter

The server configuration changed for the next purchase and we moved to a stacked solution with about 1800 servers in the bottom container and all electrical and mechanical systems in a second container on top, which is slightly shorter to fit within the datacenter’s height limits.

Double stacked container in the Chicago datacenter

The next evolutions of containers were not restricted by ISO height. The increased height allowed the same number of servers to be housed in a single taller container. Our floor space constraints restricted the width and length to be within ISO limits, but we had enough headroom for a taller container. We did however have to factor in highway overhead clearing for the trucks we were using to ship these containers to our facility.

The increased height of containers allow the same number of servers to be housed in a single unit

Through this experience we learned several lessons. First, the containers allowed us to bring in a large number of pre-assembled and pre-installed “contained” computing capacity to be connected into our infrastructure within hours after connecting power, network fiber, and water connection. The energy efficiency came from containment of air flow in a restricted volumetric space. The ISO dimensions were a convenience with respect to transportation and use of readily available lifting apparatus, but they placed constraints on the container design. We moved away from strict adherence to ISO dimensions over time. The 8-foot width of ISO containers limited how the server racks could be positioned and serviced.

Our experiments with free air cooling showed that we could run our servers at higher inlet temperatures and use outside air in most locations instead of using water side economization. With the addition of adiabatic cooling, the temperature range could be extended by 10 to 15 degrees. Our Generation 4 design deviated from ISO dimensions to allow more design flexibility. For example, the floor space was expanded to allow a larger footprint in our Quincy datacenter. Our free air cooled Information Technology Pre-Assembled Components (ITPACs) came in four units that were assembled on site (some assembly required J) before they could be connected to power, water, and network.

Modular PAC units being prepared for final connections in Quincy, WA datacenter

Modularity was a design goal for us, but modularity does not imply containers, and containers do not necessarily imply ISO compliance! We stopped using the term container and coined ITPACs to denote our new IT Pre-Assembled Components.

In recent discussions at several datacenter conferences, there is still some confusion about what modularity means. To some it means containers. The dictionary defines it as “the use of individually distinct functional units” or “constructed with standardized units or dimensions for flexibility”. A modular datacenter consists of smaller building blocks that can be replicated to deliver additional capacity. The form factor and size (power and floor space) can vary greatly. Our Gen 4 datacenter designs feature modular buildings that function as “colocation rooms” that can be populated with servers as needed, while the ITPACs are shipped fully populated with servers. Both are free air cooled in selected facility locations. This is the next step in our pursuit of higher energy efficiency. The physical form factor and technologies will evolve over time. ISO containers are a thing of the past; they served a good purpose at first but are not as relevant today.

The use of containers and ITPACs are another form factor for housing IT infrastructure. They are not tied any specific hardware or software configuration. IT should not be misconstrued as a cloud services appliance! Any configuration that resides inside a container or ITPAC can be installed and operated inside a colocation room, but the costs parameter will be different.

To “coin” an old car commercial phrase from the 1980s, “this not your father’s datacenter”!

Microsoft’s Quincy, WA datacenters: Gen 2 datacenter on left with Gen 4 modular facility on bottom-right side

\db

The Disappearing Datacenter….

msdcblog — Wed, 20 Jul 2011 14:00:00 GMT

By Christian Belady
General Manager of Datacenter Advanced Development

Sustainability is an interesting thing. It really means so many things to so many people but I really like the way it is defined in Wikipedia; “Sustainability is the capacity to endure.” Clearly, when taken in the context of the environment we are as an industry taking steps to try to be sustainable. If you look at the executive brief we are releasing today and the Microsoft.com story on “Microsoft’s Quest for Greater Efficiency in the Cloud”, I talk about the metrics that have been introduced recently by the Green Grid such as CUE and WUE to add to the already globally recognized metric of PUE. These are really great metrics to improve sustainability in our industry segment and as you all know I am a huge supporter of these metrics. However, the problem long term with these metrics is that they only optimize the datacenter piece in the complete ecosystem. This is why over the past couple years, my main focus (and the focus of my new team) has been looking at the whole cloud ecosystem as an opportunity for optimization. How does that change how we look at things? How can we integrate across what have been different industry segments to achieve greater efficiencies and an even more sustainable cloud ecosystem than we could have ever imagined? For example, I introduce the idea of data plants, which is the integration of power plants and datacenters. This concept can substantially improve efficiency by eliminating the need for transmission lines, substations, and transformers (as well as the associated voltage losses) that we see in today’s power distribution ecosystem. With data plants we distribute a form of energy called data in a completely different grid called the network. Looking at it this way, we are essentially taking another step in the evolution of refining the energy being distributed.

In the 1800s – coal was the primary form of energy for industry
In the 1900s – electricity (a processed form of coal) was the form of energy for industry
In the 2000s – data (a processed form of electricity) will be the form energy for industry

The point is that if we look at data as a form of energy, how does that change what we are doing today? When I talk about the disappearing datacenter, what I really mean is that it will disappear as we know it through integration and drive unprecedented levels of efficiency gain….and this can only be done at the scale of the cloud.

//cb

How Big is the Datacenter Construction Business?

msdcblog — Mon, 14 Mar 2011 17:37:00 GMT

By Christian Belady
General Manager, Data Center Research, Global Foundation Services

As the demand for cloud services continues to grow, there are many companies in the datacenter industry (particularly start-ups) that have met with me to discuss their businesses. Their proposals usually included huge expectations on what they think can be generated from revenue opportunities in constructing datacenters. It seems that most of the time, these vendors assume that their niche business projections will have great impact, when in reality they are relatively small to the actual market opportunity. Having worked in the datacenter industry for a long time, I generally do my best to try to help them understand that they are over estimating the market opportunity, but when they pressed for the numbers to back up that claim, I didn’t have any.

Figure 8. Year-over-year datacenter construction market size

My responses in the past were based off my professional observations, that even though we have seen the datacenter power footprint double every five years, to me it didn’t necessarily mean that the datacenter construction business would follow that same trajectory. Datacenter designs are emerging today that drive cost per megawatt to half of what was traditionally expected. As a result, one would expect that the actual market size (in dollars) may in fact not be growing as fast as we think.

As usual, my curiosity got the best of me and I tried to find some research that may have done on the topic: How big is the annual construction opportunity for datacenters?

Since I could not find anything, I decided to try to estimate the annual new datacenter construction market size using existing data already available in the industry. My calculations resulted in a projection out to 2020 that I believe is fairly representative of what we can expect. In addition, I also made the assumption that the industry will likely move from an average construction cost of $15 million per megawatt to something more like $6 million per megawatt in five years resulting in a flattening of year-over-year growth. The resulting white paper is titled Projecting Annual New Datacenter Construction Market Size. It also shares the steps I took to generate these projections and estimates that the United States and global datacenter markets will grow 50 percent or so by 2020 to $18 billion and $78 billion, respectively. These estimates were derived by combining my own data with the great work of my good friend Jonathon Koomey, and some of the initial Uptime Institute’s work on datacenter costs.

Figure 8 (inserted above) captures the conclusion of the paper, however I would urge you to read the work to understand all of the underlying assumptions and how I ended up with these results.

It is my hope that this paper will serve as an aid for industry businesses and venture capital firms to better understand and calculate the approximate size of the datacenter construction business that they can realistically capture. In addition, it is important to note, that this is a “first order estimate.” The purpose of the paper will also hopefully seed our industry for a dynamic debate and discussion to further inform in this area. Please send your thoughts and feedback on the paper and as always, I look forward to the dynamic discussion with my industry friends!

//CB

Offering Support to Help Victims of the Japanese Earthquake

msdcblog — Sat, 12 Mar 2011 01:45:38 GMT

By Charlie McNerney, GM, Business Management, GFS

On Friday, March 11th, 2011, a magnitude 8.9 earthquake hit Japan followed by a series of tsunamis, causing widespread damage. In response, Microsoft activated its Disaster Response protocol. We are currently accounting for all our employees and assessing all of our facilities for any impact.

We are saddened by the devastation caused by the earthquake in Japan. We are reaching out to our customers and partners to conduct impact assessments, and we are providing those impacted by the earthquake with free incident support to help get their operations back up and running. There has been no disruption to our cloud based and hosted services, and we continue to monitor the situation closely. Microsoft also has a cloud-based disaster response communications portal running on Windows Azure that is available to governments and nonprofits to use for communication between agencies and with citizens.

More information on Microsoft’s response to the disaster is on our Microsoft Corporate Citizenship website which will be updated with any new developments.
News and Resources:

• Search Bing for the latest quake information
• Get the latest news from MSNBC
• See updated images
• Watch related videos
• Get breaking news on Twitter

//cm

Microsoft’s Cloud Gets Even Greener in Ireland

msdcblog — Tue, 22 Feb 2011 16:00:00 GMT

By Kevin Timmons,
General Manager of Datacenter Services

I’m pleased to announce that we are continuing with our strong global environmental sustainability efforts in our cloud infrastructure by signing a contract to purchase renewable wind power from a local Irish energy provider for our Dublin datacenter.

Main entrance to Microsoft’s Dublin Datacenter

In addition to utilizing this new source of energy, our Dublin datacenter incorporates a free air cooling design which provides a dramatic reduction in power consumption. The removal of traditional datacenter cooling systems, used in the industry today, also negates the need for a chiller plant, allowing us to avoid the consumption of around 18 million liters of water each month that a traditional datacenter of its size would typically use.

Our approach to environmental sustainability in Dublin is just one more example of our overall holistic approach to designing and building datacenters around the world. Our strategy focuses on innovating locally and designing facilities that take advantage of the region’s natural environmental benefits where possible.

Our team is continuing to evolve our modular building strategies for even more energy efficiency and we will continue to share our best practices with others in the industry.

//Kev

Datacenter Efficiency

msdcblog — Tue, 01 Feb 2011 22:42:00 GMT

Datacenter Efficiency
By Dileep Bhandarkar, Ph.D.
Distinguished Engineer

Global Foundation Services, Microsoft
Datacenters have captured the attention of the computing industry over the last few years, fueled mainly by the tremendous growth in cloud computing. Companies need to make sure these facilities and IT equipment operate as efficiently as possible. In a few months, I will have completed my first 100 (binary!) years at Microsoft. My initial work was on rightsizing and optimizing the servers that we purchased for our workloads. As I interacted with my colleagues in Datacenter design and operations, I realized that we had a tremendous opportunity to work together on server and datacenter designs. Our work on containers in 2008 convinced us that we had to think more holistically and optimize the entire solution.

Today, I would like to share a Datacenter Efficiency strategy brief that we recently published for our customers that covers the top drivers of datacenter total cost of ownership (TCO) and return on investment (ROI), how Microsoft “right-sizes” servers and optimizes power and performance, and shares our roadmap for future datacenter design. You can also watch a short video where I explain how Microsoft’s approach to thinking of the datacenter as the server and right-sizing servers to production environments results in lower costs and greater efficiencies.

This approach to achieving energy efficiency becomes even more important as our cloud infrastructure and services scale grows and cost competiveness is a key to success. We now holistically design our servers and datacenters together. A good example of some of the new direction we are moving in can be seen in the removal of the server fans from our latest free air- cooled ITPACs. To the casual observer this may seem like an increase in PUE because the IT power was reduced and the infrastructure power was increased. But the bottom line is that we get the same amount of computation for less total energy consumption!

Several years ago, we moved towards evaluating the Total Cost of Ownership in our server purchase decisions. Higher server power not only drives higher energy consumption costs, but it also uses a larger share of the expensive power delivery and cooling infrastructure. We have seen several examples where it has been prudent to increase the acquisition cost slightly if it results in overall savings in power related costs. Additionally, we continue to invest in datacenter designs that reduce our capital investment and operating costs.

I am especially excited about our new datacenter designs that are modular and more efficient. Our container experience has taught us that it is all about “containment” and not the containers themselves. For example, our Dublin datacenter uses outside air and hot aisle containment in “PODs” to achieve higher energy efficiency. We are continuing to evolve that concept in future datacenter designs that drive will drive even greater efficiencies.

We will continue to share our experiences and best practices as we into the next phase of improved datacenter efficiency.

\db

The paper and video can be found on GFS’ website >>

The Gravitational Pull Was Too Strong

msdcblog — Thu, 27 Jan 2011 17:00:00 GMT

By Christian Belady
General Manager, Data Center Research, Global Foundation Services

Many of you may have already heard that I have change roles yet again in Microsoft. It was only 11 months ago that I had moved to Microsoft Research to pursue ground-breaking work in the Extreme Computing Work (Apples Don’t Fall Far from the Tree). In that time, I experienced the great opportunity of helping to set an organizational vision and build a team of Hardware Architects and Engineers, who I believe are second to none. It was a significant experience given to me by Dan Reed the CVP of XCG, which I am very grateful for. However, while their team’s ship has set sail on a course that I believe will transform the industry five to 10 years out, I realized that growing up professionally in a product development ecosystem most of my career, I do need more of an instant gratification. Combined with that need and the growing business needs of the Global Foundation Services (GFS) team, I am thrilled to say that I will be going back to my roots in Microsoft to be the GM of Datacenter Research (DCR), reporting to Kevin Timmons.

DCR will be an advanced development lab (versus a traditional research lab), whose horizon is one step beyond the datacenters we have on the drawing board today. I am blessed to have yet another great opportunity.

So what was the defining moment that made me decide on yet another move? In October, I took a field trip with my team to take a look at the progress of our new Quincy, Washington modular facility. Upon seeing the facility, my first reaction was euphoric. The facility looked exactly like we had been conceptualizing it on a white board in my previous stint in GFS. Wow, it was very gratifying knowing that I was a part of the team that envisioned the design. Soon after we nailed the concept, I had moved on while the rest of the team experienced the satisfaction of being able to realize the completion of the engineering and execution of the airside, economized, modular design. They are the ones that deserve all of the credit for a truly break through datacenter…they are a great team. And I wanted to be part of that team again, it was that simple. I also have complete respect and trust for the GFS management team in their values and strategic direction. So when the leadership in GFS offered the opportunity, there was no question in my mind on how to proceed. So here I am again, excited about what the future will bring and like I said “The Gravitational Pull Was Too Strong” and the apple returned to the tree!

Finally, just as you may have heard us say before, whatever you see today in our datacenters, you’re not likely to see tomorrow, as we are a company that is constantly evolving and innovating. As a result, the opportunities are endless for its datacenter engineers and services operations team, and for the customers that we serve. Man, I love this company!

/cb

Shedding Light on Our New Cloud Farms

msdcblog — Tue, 04 Jan 2011 15:00:00 GMT

By Kevin Timmons,
General Manager of Datacenter Services

It’s an exciting time for Microsoft’s datacenter program. In addition to operating one of the largest global datacenter footprints in the industry, we have been super busy working on multiple next-generation, modular facilities that are in various phases of construction.

One of our most innovative new datacenters is set to open in Quincy, WA in early 2011 and incorporates key learnings from award-winning facilities that Microsoft opened last year in Chicago and Dublin. The Dublin facility uses server PODs and outside air economization to cool the servers, which significantly reduces cooling expense and infrastructure costs. We took a slightly different approach with our Chicago datacenter which utilizes water-side economization for cooling and improves scalability by using IT Pre-Assembled Components (ITPACs.) An ITPAC is a pre-manufactured, fully-assembled module that can be built with a focus on sustainable materials such as steel and aluminum and can house as little as 400 servers and as many as 2,000 servers, significantly increasing flexibility and scalability. You can learn more about our ITPACs by going to the ITPAC video

The expansion in Quincy takes these ideas a step further by extending the flexibility of PACs across the entire facility using modular “building blocks” for electrical, mechanical, server and security subsystems. This increase in flexibility enables us to even better support the needs of what can often be a very unpredictable online business and allows us to build datacenters incrementally as capacity grows. Our modular design enables us to build a facility in significantly less time while reducing capital costs by an average of 50 to 60 percent over the lifetime of the project.

View Microsoft's Modular Datacenter slide deck >>

When Phase 1 opens in Quincy it will be located adjacent to our existing 500,000-square-foot facility. However, the new datacenter is radically different. The building will actually resemble slightly more modern versions of the tractor sheds I spent so much time around during my childhood in rural Illinois.

Tractor shed in my home town of Mt. Pulaski, IL

The building’s utilitarian appearance belies its many hidden innovations. The structure is virtually transparent to ambient outdoor conditions, allowing us to essentially place our servers and storage outside in the cool air while still protecting it from the elements. The interior layout is specifically designed to allow us to further innovate in the ways that we deploy equipment in future phases of the project. And, like any good barn, the protective shell serves to keep out critters and tumbleweeds.

Additional phases have been planned for the Quincy site and will be built based on demand. Those phases will incorporate even more cutting-edge methods to deploy servers and storage in ways that have never been seen before in the industry.

We will open other modular datacenters later in 2011 in Virginia and Iowa and I’ll be sharing more information about those facilities at that time. Our modular approach to design and construction with these facilities will allow us to substantially lower cost per megawatt to build and run our datacenters while significantly reducing time to market. This is the holy grail for most datacenter professionals…. fast, cheap and reliable – what more could you ask for?

We’ve been sharing our research and best practices around modularity with our partners and others in the industry for a number of years and I’m thrilled to see the industry begin moving in this direction. By sharing our learnings, we’ve helped others build more sustainable facilities and reduce our collective carbon footprint.

Stay tuned for more information about our future datacenter projects in 2011.

//Kev

Holistic Approach to Energy Efficient Datacenters

msdcblog — Mon, 02 Aug 2010 15:00:00 GMT

By Dileep Bhandarkar, Ph. D.
Distinguished Engineer
Global Foundation Services, Microsoft

A little over three years ago, I joined Microsoft to lead the hardware engineering team that helped decide which servers Microsoft would purchase to run its online services. We had just brought our first Microsoft-designed datacenter online in Quincy and were planning to use the innovations there as the foundation for our continued efforts to innovate with our future datacenters. Before the launch of our Chicago datacenter, we had separate engineering teams: one team that designed each of our datacenters and another team that designed our servers. The deployment of containers at the Chicago datacenter marked the first collaboration between the two designs teams, while setting the tone for future design innovation in Microsoft datacenters.

As the teams began working together, our different perspectives on design brought to light a variety of questions. Approaches to datacenter and server designs are very different with each group focusing on varying priorities. We found ourselves asking questions about things like container size and if the server-filled containers should be considered megaservers or their own individual datacenters? Or was a micro colocation approach a better alternative? Should we evaluate the server vendors based on their PUE claims? How prescriptive should we be with the power distribution and cooling approaches to be used?

View inside a Chicago datacenter container

After much discussion we decided to take a functional approach to answering these questions. We specified the external interface – physical dimensions, amount and type of power connection, temperature and flow rate of the water to be used for cooling, and the type of network connection. Instead of specifying exactly what we needed in the various datacenter components we were purchasing through vendors, we let the vendors design the products for us and were surprised at how different all of their proposals were. Our goal was to maximize the amount of compute capability per container at the lowest cost. We had already started to optimize the servers and specify the exact configurations with respect to processor type, memory size, and storage.

After first tackling the external interface, we then worked on optimizing the server and datacenter designs to operate more holistically. If the servers can operate reliably at higher temperatures, why not relax the cooling requirements for the datacenter? To test this, we ran a small experiment operating servers under a tent outside one of our datacenters, where we learned that the servers in the tent ran reliably. This experiment was followed by the development of the ITPAC, the next big step and focus for our Generation 4 modular datacenter work.

Our latest whitepaper, “A Holistic Approach to Energy Efficiency in Datacenters,” details the background and strategy behind our efforts to design servers and datacenters as a holistic, integrated system. As we like to say here at Microsoft, “the datacenter is the server!” We hope that by sharing our approach to energy efficiency in our datacenters we can be a part of the global efforts to improve energy efficiency in the datacenter industry around the world.

\db

The paper can be found on GFS’ web site at www.globalfoundationservices.com on the Infrastructure page here.

Dublin Datacenter Recognized by Industry as the Best European Enterprise Datacenter Facility

msdcblog — Mon, 12 Jul 2010 02:32:00 GMT

Last week, at the 6^thData Centres Europe 2010 Conference, our Dublin datacenter received a prestigious award for being the Best European Enterprise Datacenter Facility. The Dublin datacenter incorporates a free air cooling design and is one of the most energy efficient facilities Microsoft has designed and built so far. Beyond the dramatic reduction in power consumption, the removal of traditional cooling systems negates the need for a chiller plant, allowing us to avoid the typical consumption of around 18 million liters of water each month. This design shows what is possible when you innovate locally and design facilities which take advantage of the natural environmental benefits around you.

This most recent award is the third time that our Dublin datacenter has won an award over the last year. Other awards include the Datacentre Leaders Award for Innovation and the European Code of Conduct for Data Centre Sustainability Best Practice. These awards highlight our fresh approach used in Dublin where we integrated new design solutions to achieve greater efficiencies and waste reductions.

In addition to awards won by our Dublin datacenter, our Chicago datacenter also won the Green Enterprise IT Award from the Uptime Institute for the bold IT initiatives we utilized in designing the facility. The Chicago datacenter integrates new design solutions which have led to greater efficiencies and carbon waste reductions. This includes a significant reduction in water and material usage enabled by the deployment of containers and other IT innovations.

I’m extremely proud of the hard work put forth by my team in making our award-winning datacenters a reality and for everyone responsible for running the day-to-day operations of Microsoft’s datacenters. Your ingenuity in designing and operating more energy efficient datacenters that produce less waste has been recognized by the industry and I have no doubt that we’ll achieve even greater results this coming year. Great job!

Chicago Datacenter Wins Green Enterprise IT Award from Uptime Institute

msdcblog — Tue, 18 May 2010 15:18:00 GMT

Today, Microsoft won a prestigious Green Enterprise IT Award from the Uptime Institute for the bold IT initiatives we utilized in our recently opened datacenter in Chicago. The facility integrates new design solutions which have led to greater efficiencies and carbon waste reductions. This includes a significant reduction in water and material usage enabled by the deployment of containers and other IT innovations. The Uptime Institute is an organization that is focused on providing education and thought leadership for the enterprise datacenter industry and for datacenter professionals. The Green Enterprise IT Awards are the Uptime Institute's highest honor.

By incorporating containers we have reduced carbon waste from extra packaging of servers and equipment and individual transport shipping consumables such as cardboard, styrofoam, plastic, and other packaging materials. The highly innovative approach to design, construction and operations has also allowed us to fully deploy thousands of servers within a few hours, with a drastically reduced carbon footprint and great cooling efficiencies, with a reduced PUE of 1.15 - 1.22 across the facility.

Containers in the Chicago Datacenter

When we began construction of the facility in 2007, our belief in modularization and the use of containers as computing modules at huge scale was not only a stretch for our organization, but for the entire industry. Our container approach in Chicago is also being utilized by others and has helped bring much needed new ways of thinking about how to deploy capacity in shorter timeframes, while driving more engineering efficiencies in the datacenter industry. Another Green IT approach has been the use of water-side economization which enables us to cool the facility without requiring the high levels of electricity typically needed to power large chillers.

I’m particularly proud of the way that our team has continued to drive new operational sustainability technologies across the entire datacenter infrastructure and IT, and that we continue to be committed to sharing our research findings and best practices with others in the industry. By sharing our best practices, we hope others in the industry will also be able to reduce their overall carbon footprint, and ultimately help make all of us even better stewards of the environment.

//Kev

You can find out more details about Microsoft’s Chicago datacenter here

Celebrating Earth Day from the Emerald Isle

msdcblog — Tue, 20 Apr 2010 21:19:00 GMT

By John Dwyer – International Datacentre Manager

The 40th anniversary of Earth Day is a very appropriate time to reflect on the aggressive sustainability objectives that Microsoft has for the Dublin datacentre and the great progress we’ve made so far. As you will recall, the facility went into production on July 1, 2009, and with its free air cooling design is one of the most energy efficient facilities Microsoft has designed and built so far. Beyond the dramatic reduction in power consumption, the removal of traditional cooling systems negates the need for a chiller plant, allowing us to avoid the typical consumption of around 18 million litres of water each month.

Microsoft’s Dublin datacenter server Pods utilize fresh outside air year round

In the 9 months since we moved into the production phase of the facility, the two initial server rooms have seen rapid growth in the number of operational systems in place as we have replaced less efficient systems that previously served the European market. Our engineering and operations teams are also continuing to partner closely to assess and drive the implementation of new strategies for further efficiency optimizations. Our determination and commitment to running highly sustainable datacentres is as firm as it has ever been and I look forward to sharing more on our progress over the coming year.

Microsoft is working hard with our partners, customers and suppliers, as well as with governments and leading environmental organizations, to help address this critical environmental challenge. You can read more about our corporate commitments and Earth Day perspectives via our Microsoft Environment web site.

As always, I welcome your questions and feedback.

Cheers

JD/

One Small Step for Microsoft’s Cloud, Another Big Step for Sustainability

msdcblog — Tue, 23 Mar 2010 13:20:00 GMT

By Kevin Timmons

General Manager of Datacenter Services

The datacenter industry is at an inflection point. We need the ability to build out “at right scale” based on capacity needs at unprecedented price points. Oh by the way - it’s also imperative that we find ways to avoid the significant up-front costs and long lead times required to build traditional datacenters. Mission impossible, right?

Wrong. In order to address these challenges Microsoft has been moving toward more modular designs over the last couple of years, moving from purchasing individual servers to racks of servers to, most recently, entire containers filled with servers. Our most recent datacenter in Chicago used the container approach. At our datacenter in Dublin we were able to eliminate the need for chillers by using stand-alone server pods. These advances helped us become more efficient in our datacenter operations but we are continuing to look for new ways to look at the problem.

During my recent keynote speech at the Datacenter Dynamics Conference in New York (in March 2010), I shared details about the next step in Microsoft’s datacenter evolution, which is consistent with the vision that we’d previously laid out as our Generation 4 datacenter. Our plan is to move towards pre-manufacturing every part of the datacenter and assembling them onsite based on the class of service to be delivered and supported. By eliminating over-provisioning and maximizing productivity on an ongoing basis we are recognizing that each datacenter is a living organism that needs to be flexible enough to adapt quickly to meet our own demands as well as the demands of our customers.

One of our ITPAC “Proof of Concept” modules for datacenters

Our plan for the future is to have essentially everything but the concrete pad pre-manufactured and then assembled on site: the IT, mechanical and electrical components are all part of Pre-Assembled Components that we call an “ITPAC.” We actually think of the ITPACs not as containers in a traditional sense but as integrated air-handling and IT units.

The units will be assembled entirely from commercially available recyclable components such as steel and aluminum and the cooling requirements for the ITPACs will be met by more efficient means, such as a single water hose with residential levels of pressure to control ambient temperatures. The servers will be stacked in rows, sandwiched between air intake and output vents.

Using fans to create a negative pressure for cooling, ambient air can be pulled through on one side, run through the servers and exhausted out the other, with some of the air re-circulated to even the overall temperature of the unit. Think of how a radiator works in a car. No mechanical cooling units will be used and networking and power buses will run over the tops of the servers. The elimination of mechanical cooling devices should allow the Average PUE ratio (drives OPEX) for our ITPACs to be approximately 1.15 to 1.19 and the Peak PUE ratio (drives CAPEX) to be around 1.32 to 1.38, depending on the outside conditions and the number of hours of ambient air cooling in a year.

Our development team is considering a number of different sizes of ITPACs in order to make the units easily shippable, and they could contain approximately 400 to 2,500 compute servers and draw 200 to 600 kilowatts depending on the server mix between compute and storage. Another exciting development is that in our research and development, we found that with automation a single person could build one of these units (like pictured above) in only four days. Click here to watch a short video on how we build one of these ITPACs.

Once built, the units can be placed inside a large building or when equipped with outer protective panels, they can reside out in the open and be linked together to build out an entire datacenter. We believe that by utilizing this new approach, Microsoft can reduce the time it takes to ramp up new cloud computing capacity in half the time as traditional datacenter infrastructures, as well as significantly reduce the cost of the building. This gives us the flexibility to grow without having to commit to a large upfront investment for a datacenter and hope that demand shows up later.

These facilities will not be pretty and might actually resemble the barns I spent so much time around during my childhood in rural Illinois. That, combined with the fact that these facilities will be substantially lower cost per megawatt to build and substantially lower cost to run, makes it very easy to become excited about what we’re doing here.

William Gibson said it best: “The future is here, it’s just not widely distributed yet.”

Today we are taking another big step towards greater efficiency and driving down costs through a more widely distributed – modularized and commoditized datacenter. My thanks go to the thoughtful engineering approaches of our team and the close collaboration from our product, services and research teams across the company, as well as our partners–in helping us achieve these objectives. We do not publically disclose information on the partners and vendors we work with within our cloud infrastructure due to our policies to protect these high security facilities. However, we have shared our research and development information for these ITPACs and other Generation 4 technologies with key partners over the past couple of years and many of them are incorporating this information into their product offerings for customers today.

Few companies have the capability of tapping into $9.5 billion dollars of Research and Development that Microsoft supports (company-wide) per year. And we hope that by sharing our key learnings and best practices that we can help the industry, as a whole, work together to drive greater efficiencies through our cloud services infrastructures and collectively reduce our carbon wastes.

/Kev

You can access more of our best practices around environmental sustainability, cloud security, etc. in published papers and blogs available on our web site at http://www.globalfoundationservices.com.

Apples Don’t Fall Far from the Tree

msdcblog — Thu, 04 Mar 2010 01:36:00 GMT

By Christian Belady

Director of Hardware Architecture, Microsoft Research

Some folks have already heard that I am changing roles in Microsoft, but the reality is that just like the evolution with our data centers, my career is going through a similar evolution. And that’s the way I like it….while I love what I do, I continuously like to take on new challenges and evolve. In any case, before I dive into what I will be doing, I thought it would be great to reflect on my almost three years at Microsoft.

When I first hired into Microsoft, the company had already recognized the need for monitoring and measuring (using our SCRY technology) to improve operational efficiency in our data centers. They had already adopted my beloved PUE metric and tracked it daily in all of its data centers, so I knew Microsoft was a progressive company and a place I wanted to be. While Microsoft was already driving improvements in efficiency as a result of this, many in our industry were still in a “wait and see” mode. The satisfying thing is that in less than three years the industry has pretty much accepted that efficiency and PUE are pretty much a universal best practice and metric to adopt. Also, that monitoring and measurement is a key enabler to drive the right behavior. I have to believe that Microsoft and our data center services team played a role in making that happen with the effort we all spent on evangelizing the cause. We have invested our time in supporting the EU Code of Conduct, The Green Grid, Climate Savers, EPA Energy Star for data centers program, and many other industry organizations that have followed suit in adopting common efficiency metrics and approaches. We all should feel pretty good about that. Now looking inside of Microsoft, I feel proud of what we as a team accomplished since I joined…you can argue that our data center strategy during this period is likely the most dynamic seen in the industry to date; the most memorable examples (for me) are the container computing scale approach in Chicago, server PODs and air side economization in Dublin, our Gen 4 vision, and perhaps even our “tent city” experiment to demonstrate extreme economization. In my mind these were all game changers and showed that there is much room for innovation in an industry where change has historically been rather slow.

But even with all of this change, I see there is even more opportunity now then there was when I started at Microsoft almost three years ago. Cloud computing has made mining and developing the “right” opportunities even that much more important. We need to think about how we tie together the complete ecosystem of the software stack, the IT, the data center and the grid today and what efficiencies we can drive from our research and development for the future. For those of you that know me – this is the kind of opportunity that makes me salivate. There aren’t many people around tasked with this kind of challenge and this is the opportunity I have been given in the evolution of my career at Microsoft. This week I begin tackling these projects within the Microsoft Research group in team called the Extreme Computing Group.

This is exciting in itself, but what really gets me “charged” is the opportunity I have now to work between my former group Global Foundation Services (that drives the current Microsoft cloud infrastructure) and my new group Microsoft Research (MSR). Taking the best practices from what we have learned with our current and future Gen 4 data centers and combining them with the resources of one of the best research organizations in the world (MSR), I am convinced that many new and exciting things will come. And best of all, I am lucky to be right smack in the middle of it and will still be working closely with the teams driving the hardware architecture for the cloud today and in the future. So actually, I am really not leaving GFS but rather extending the reach of GFS into the future. Who can ask for a better opportunity….man I love this company!

Thanks,

/cb