Dates Provide A Clue
But the chart included dates, which allows us to do some math to fill in the blanks. In a presentation in November 2009, Facebook vice president of technology Jeff Rothschild disclosed that the company had more than 30,000 servers. Cook’s chart shows a brief plateau in Facebook’s server growth at about that time, followed by a sharp upward spike in the growth line through the first quarter of 2010 that effectively doubles the total number of servers.

That suggests that Facebook now has 60,000 or more servers. The sharp acceleration in Facebook’s server growth in late 2009 also helps explain the company’s move to lease large chunks of data center space in northern Virginia and Silicon Valley in March. The growth spurt occurred after Facebook announced plans to construct its own data center in Prineville, Oregon.

That places Facebook among the largest Internet companies that have publicly discussed their server growth, but still well behind Intel, which has more than 100,000 servers. See Who Has The Most Web Servers for more data on the largest Internet infrastructures.

Why so many servers? Facebook now has more than 400 million active users, including more than 200 million who use the service every day. Here are some other data points from Cook’s presentation and a talk last Thursday at Structure 2010 by Facebook’s Jonathan Heiliger.

• Users spend more than 16 billion minutes on Facebook each day
• Every week users share more than 6 billion pieces of content, including status updates, photos and notes.
• Each month more than 3 billion photos are uploaded to Facebook.
• Users view more than 1 million photos every second
• Facebook’s servers perform more than 50 million operations per second, primarily between the caching tier and web servers
• More than 1 million web sites have implemented features of Facebook Connect

Cloud computing is widely acknowledged as the major force driving IT into the future. International Data Corporation (IDC) has predicted that spending on cloud services is set to increase from $16.2 billion in 2008 to $42 billion by 2012. (1) But as the cloud revolution continues, companies need to consider how they’re going to update their IT infrastructures to prepare for this growth.

Several companies have already realized the benefits of colocating to a carrier-neutral data center to meet these needs. The environment found in such data centers is ideal for cloud services because it provides the opportunity to scale to match fluctuating demand, as well as the power, connectivity and secure processes needed to ensure reliable performance and continuous availability. It also provides significant economies of scale, which helps keep costs to a minimum at this critical stage in the development of cloud services. An additional benefit is the drastic cost savings in CapEx and OpEx, which can quickly and easily reach $25 million for the data center build alone.

With this in mind, there are several key factors for having an ideal environment within which cloud computing services can grow and prosper:

1. Limitless Scalability

Cloud computing demands higher levels of faster scalability than previous delivery models. Bandwidth and processing power need to be instantly available for surges in demand, with the added ability to reduce resources when peaks in traffic have passed. By outsourcing to a large data center, companies do not have to worry about over-provisioning for spikes in demand, therefore avoiding issues of under-utilization. Using a third-party data center also prevents the risk of running out of capacity for such heightened demands.

2. Physical and Virtual Security

Although there are many benefits of moving to the cloud, there are also some risks. The Ethernet-based cloud is not impenetrable or fail-safe and is certainly not immune to data loss. Organizations must identify operational and security risks associated with the cloud, namely data security, integrity and privacy, so as to better choose a solution that addresses these concerns.

Large data center environments are ideally suited to secure delivery of cloud applications, such as Storage-as-a-Service and Software-as-a-Service. This is due to the robust nature of the data center’s infrastructure and the inherent need for high-quality, efficient and up-to-the-minute technologies and hardware. Larger facilities excel in both physical and data security, with multiple security layers and fail-safes, as well as back-up and recovery systems that protect against data loss.

3. High-density Power

While most corporate servers average 15 percent utilization, virtualized servers can run at 60 to 80 percent. As a result, high-density power has become a near necessity. High-density power availability enables cloud service providers to deploy the very latest and most efficient equipment while minimizing the space required and, therefore, the cost to the business.

4. Broad Connectivity Choices

As virtualized infrastructures and the demand for “anywhere, anytime” access continue to increase, connectivity has become as important as processing power for users, meaning a company’s implementation of cloud computing will succeed or fail based on the quality of the end-user connection. Maximum bandwidths and multiple connectivity options will drive adoption of the pay-per-use model, and anything less will have the reverse effect. A carrier-neutral data center can provide the widest possible range of connectivity options, and with increased choice, comes more opportunities for lower costs.

5. Low Cloud Costs

Having economies of scale is the greatest benefit of scaling up the infrastructure supporting the cloud. Large data centers drive down the cost of hosting cloud computing infrastructures, multiplying the economies generated by the cloud delivery model and protecting profit margins for customers.

A recent University of California study showed that, when looking at storage, networking and processing costs, very large data centers reduce costs by five to seven times. (2) Additionally, the build-out cost per square-foot reduces data center capital costs by up to 60 percent, according to Tier 1 Research. (3)

Outsourcing cloud services also helps companies that cannot afford to build or continue to expand their own data center(s), thereby reducing their CapEx and OpEx. Facility costs are growing rapidly, and according to the Uptime Institute, the true costs of running a server are often four to five times the cost of the server alone over a five-to-10-year lifetime. (4) Using a third-party data center equates to great savings in both cost and administration for the business.

6. Dedicated Community

Moving the capabilities beyond simple colocation, some data center providers have built up specific “cloud communities” – or communities of interest – of organizations that benefit from colocating in close proximity to markets and within the same colocation services provider. For instance, a data center community built within the financial services sector, where speed of trading between businesses is crucial, will help companies gain the benefits of speed to market and proximity to financial markets.

Additionally, organizations within such communities gain direct access to financial exchanges, brokers, market data suppliers, and a large range of managed services providers and ISVs, many of which connect and use a common infrastructure and/or applications. This type of community approach also applies to substantial areas of interest among digital media and service providers, carriers and Internet exchange and enterprise sectors.

The Maturing Cloud

Ultimately, what will be vital to the success of cloud computing, and to those wanting to take full advantage of its benefits, is a reliable infrastructure that can scale, while simultaneously providing the required connectivity and security.

Colocation data centers are growing because they satisfy the basic need to access useful information, quickly, at lowest cost and minimal risk. More and more companies are realizing the benefits of having a portion of their data processing, storage and networking needs in a third-party facility to help ensure their operations can continue free of disruption.

Cloud computing is still an evolving paradigm that will take many years to fully mature. However, with improvements in understanding and awareness of security, infrastructure and regulatory procedures, and with solid infrastructure foundations in place, the cloud looks set to continue its path of growth and adoption.

This centre was inaugurated by Mr. Khalid Kalban, Managing Director and Chief Executive Officer of Dubai Investments. The data centre will provide clients with access to a fully-redundant IT infrastructure with 24/7 onsite security and technical support, in addition to a comprehensive range of value-added service offerings.

Dubai Investments also announced that it will create a subsidiary company to manage the data center collocation and managed services business, in order to efficiently meet the growing demand for these services from various companies in Dubai Investments Park. The new subsidiary will form strategic partnerships with converged network providers for connectivity, and with enterprise solution providers and information security vendors for delivering software and security as a service to the companies.

Shukri Al Mehairi, General Manager, Dubai Investments Industries, said, “This state-of-the-art data center will serve as a robust infrastructure for IT operations of DI subsidiaries and companies in Dubai Investments Park. Companies today rely extensively on its information systems to efficiently run operations, making it imperative to have the support of a dedicated facility that hosts mission critical computer systems.”

“Our new data center offers a scalable, no single-point-of-failure infrastructure. Moreover, the server consolidation, storage consolidation and in-row cooling contribute to increased power saving, making it a green data center,” added Kalban.

Manoharan Kothandraman, IT Advisor, Dubai Investments, said, “The data center and managed services will offer companies in Dubai Investments Park a full-fledged IT setup on a monthly service fee. We are certain that companies will soon see a significant increase in business value without major capital investments and huge spending on internal IT resources.”

The data center will host enterprise IT solutions like ERP (Enterprise Resource Planning), e-mail, CRM (Customer Relationship Management) and Human Resource Management Systems, and will manage information security solutions like firewall, anti-virus, anti-spam, VPN (Virtual Private Network) and web filter. This will benefit the subsidiaries of Dubai Investments as well as the companies in Dubai Investments Park. Moreover, apart from hosting websites of client companies, the new data center will offer collocation facilities for companies that want to manage their servers independently.

Certified Data Center Facilities Operations Manager Training (CDFOM )

Happening only once in 2010 in Burj-al-Arab, Dubai – UAE

No. 1 worldwide accredited data center training at Burj-Al-Arab, Dubai

Certified Data Center Professional (CDCP) :  June 20th – 21st, 2010

Certified Data Center Facilities Operations Manager (CDFOM):  June 22nd, 23rd  & 24th, 2010,

Certified Data Center Professional (CDCP)  – 14^th & 15^th March

Certified Data Center Specialist       (CDCS)    – 16^th & 17^th 18^th March

The new product will be launched on January 27, during ‘Data Center Tops’, a day-long conference and technology road-show on advancements, standards, and innovations in mission-critical and high-availability data centers.

TechXact, a global leader and provider of state-of-the-art total data center solutions, is organizing the event to enable participants observe, hear and interact with the world’s leading data center manufacturers, technology suppliers, and service providers as they showcase and share the latest advancements, developments, standards, best practices and concepts.

The company, which has been developing cutting-edge technologies for over years, is bringing the unique product to the region for the first time. The product features one of the world’s most advanced technologies in fire prevention. The new system as its name suggests aims at preventing the chances of fire rather than curtailing or limiting the damage after an incident.

A live demonstration will be held following the unveiling of the product. Speaking during the press conference today to announce the conference details Mr. Mehdi Paryavi, President of TechXact, said: “When we say ‘next-generation’ we are serious about it.  For example everyone has been talking about fire suppression till now. We believe in our type of business killing fire after it has already began is just not enough and in many instances too late when you calculate downtime, SLAs, facility costs, etc…  Hence, we are bringing the notion of fire prevention.”

TechXact also plans to launch several other systems that can enable companies enhance its data centre management, compliance and energy efficiency.

According to Paryavi, the event is a must attend for senior officials of all companies with data to be stored, managed and maintained. “Techxact will introduce some of the most advanced systems currently available in the market. With green initiatives taking centerstage, and given today’s business and environmental requirement, we are emphasizing on green solutions for the market. The conference will also focus on innovations being introduced in this field and how companies can make the best use of available technologies,” said Paryavi.

TechXact has been promoting advanced solutions working with customers and partners in the Middle East, CIS, Africa, Europe, Asia and America regions. Among its leading clients in the region are civil and security establishments such as Dubai e-Goverment and the Dubai Police and leading companies and conglomerates such as Emaar Properties, GatewayGulf, Al Jabber Group, Al Tayer Group, ADNOC, Inteltec and French Connection among others.

Luai A. Jezawi, Director of Business Development (MEA) at techxact said:

“Millions are being invested by organizations in business continuity effort and high availability of their mission critical data centers. It could cost double the investment to minimize risk-window of downtime in a data center from 4 hours (per year) to 40 minutes! Fire incidents – even with traditional Fire Suppression functions in place – could potentially result in several hours to several days (maybe weeks) of IT Services outage. This is quite serious; it simply means that sizeable investments and efforts in quality assurance & availability could be wasted; not to mention severe hard dollar and soft dollar losses. Therefore, it is very crucial to protect the system from going offline, especially from fire accidents which can, now, be effectively prevented. We have tied up with some of the leading players in the industry to develop our latest product. Most of the components used are of very high grade and is aimed at achieving the best results.”

One of the components – Wagner’s Oxyreduct – he said is a system based on preventing fire by controlling the Oxygen and Nitrogen balance in the environment we desire to protect. “The Oxyreduct system makes it ‘physically’ impossible for fire to ignite to begin with; by reducing the level of Oxygen by 5% and substituting the balance with Nitrogen through regulated supply into the closed area that is being protected,” he added. Should highlight it has been tested and certified for safety – so the protected environment is surely safe for humans.

The second major risk factor for Data Center Downtime are electrical faults. Therefore TechXact provides Wagner’s advanced TITANUS Aspirating smoke detecting technology. The system is very advanced and sensitive and uses intelligent signal processing technology with a computer chip which has 6,000 fire development crafts stored to avoid false alarms.

According to Jezawi, by investing in such advanced systems companies can save crucial time and cost otherwise lost as a result of an outage. “The damages caused by a downtime can be tremendous. The loss in revenue per hour is calculated at $2,817,000 per hour when it comes to the energy sector, and $2,066,000 for the telecommunications sector. When it comes to the IT sector it can go up to 1,344,000 per hour. Hence it is of utmost importance for companies to invest in reliability, availability and quality assurance” he said quoting a research by META Group.

Also present during the press conference was Thomas Schaedlich, General Manager of Wagner Middle East FZE.

The exclusive event to be held at Emirates Palace in Abu Dhabi on January 27, 2010  is being tagged as a must attend for all whose core interest is establishing or maintaining the lead in the field of data center investment, ownership, management, operation and services.

The conference will feature several key lectures by industry experts focusing on issues ranging from new trends and innovations within the industry, to standards, compliances, best practices and totally green solutions for data centers.

Mr. Mehdi Paryavi, President of TechXact, said:  “Our objective is to assist organizations in bridging the gap between the system generally being followed and the latest technological advancement and to achieve industry’s best practices and familiarize them with new trends that can actually enhance ROI and TCO. Participants can observe, hear and interact with the world’s leading data center manufacturers, technology suppliers, and service providers.”

Datacenter Tops

Exclusive Datacenter Technology Event for the Elite Corporate and Government executives.

Where Data Center “Top” owners, operators, providers and vendors gather in one place, at one time!

The world’s leading data center manufacturers, technology and service providers, will be showcasing the latest technology advancements, developments, standards, practices and concepts gathered under one roof in one of the world’s most prestigious venues.

Data Center TOPS is an essential day not to be missed to maintain the lead in the field of data center investment, ownership, management, operation and services.

The objective is to observe, hear and interact with the world’s leading data center manufacturers, technology suppliers, and service providers as they showcase and share the latest advancements, developments, standards, best practices and concepts in an exclusive atmosphere.

Throughout the day we will be focusing on the innovations being introduced to the marketplace to ensure that IT and data center management CAN go green, implement compliances, standards, benchmarks, education, perform audits and qualify for certifications, feedback on compartmentalized data centers, innovative cooling, EMS / BMS, high-end security systems and control, revolutionary fire prevention technologies and precision civil works, to become an integral part of organizations’ quest in obtaining necessary efficiencies, corporate objectives all inline with government rulings.

Making the right moves to implement these latest technological systems and increase understanding and knowledge as to the difference these will make to an organization – providing a cutting-edge in the industry and a competitive-edge in today’s competitive markets.

All these veterans in the industry and senior experts will be gathered under one roof in one of the world’s most prestigious venues – Emirates Palace in Abu Dhabi. Attendance is all-inclusive of this grand luxury hotel’s breakfast, sumptuous international lunch buffet and coffee breaks.

The exclusive free-to-attend event for VIPs, presents a unique experience to interface and network with our superb international panel of speakers – all senior specialists in their own fields – together with your like-minded peers who identify the need for the continuing improvement in data center efficiency, streamlining and its effect on profitability – and the environment!

“For data center geeks, we expect our Buffalo data center design will have an annualized average PUE (power usage effectiveness) of 1.1 or better,” said Filo. That would challenge the most efficient facilities built by Google (GOOG), which has an average PUE of 1.16 for its six company-built data centers, with one facility currently running at a PUE of 1.12.

With its Lockport design, Yahoo is joining Google on the cutting-edge of energy efficiency by running a data center without chillers, which provide refrigerated water for cooling systesm and are among the most energy-intensive components of a data center.

In adopting a design that is less reliant on chillers, Yahoo is following the lead of Google (GOOG), which recently built a data center in Belgium that operates without any chillers. Google executives expect the Belgium facility will soon be the company’s most efficient, and may push its PUE below 1.1.

Plans filed today with the town of Lockport indicate that the Yahoo Computing Coops will be metal prefabricated structures measuring 120 feet by 60 feet. The company plans to use five of these structures in its Lockport complex. Each of the coops will have louvers built into the side of the building to allow cold air to enter the computing area. The complex is being laid out to take advantage of prevailing wind direction, which would blow directly into the louver system.

Each coop also has a peaked roof, with a “penthouse” running along the top of the structure to manage the release of  waste heat from the hot aisle, according to the Buffalo News.

Site plans suggest the coops are designed to support 5 megawatts of computing load.  The facility is designed to be completed in two phases, with a first phase using 10 megawatts of power and a second phase adding 15 megawatts.

Filko said the NewYork project is the next step in a long-term process to improve Yahoo’s data center energy efficiency.

“Today we’re committing to reduce the carbon intensity of our data centers by at least 40% by 2014,” Filo wrote. “In other words, we’ll decrease our average electricity consumption and greenhouse gas emissions from our data centers around the world. We’ll get there through a combination of innovative data center design, improving how we utilize our servers, cloud computing, and locating our data centers in areas where cleaner energy is available.”

Google (GOOG) has begun operating a data center in Belgium that has no chillers to support its cooling systems, a strategy that will improve its energy efficiency while making local weather forecasting a larger factor in its data center management.

Chillers, which are used to refrigerate water, are widely used in data center cooling systems but require a large amount of electricity to operate. With the growing focus on power costs, many data centers are reducing their reliance on chillers to improve the energy efficiency of their facilities.

This has boosted adoption of “free cooling,” the use of fresh air from outside the data center to support the cooling systems. This approach allows data centers to use outside air when the temperature is cool, while falling back on chillers on warmer days.

Google has taken the strategy to the next level. Rather than using chillers part-time, the company has eliminated them entirely in its data center near Saint-Ghislain, Belgium, which began operating in late 2008 and also features an on-site water purification facility that allows it to use water from a nearby industrial canal rather than a municipal water utility.

Year-Round Free Cooling
The climate in Belgium will support free cooling almost year-round, according to Google engineers, with temperatures rising above the acceptable range for free cooling about seven days per year on average. The average temperature in Brussels during summer reaches 66 to 71 degrees, while Google maintains its data centers at temperatures above 80 degrees.

So what happens if the weather gets hot? On those days, Google says it will turn off equipment as needed in Belgium and shift computing load to other data centers. This approach is made possible by the scope of the company’s global network of data centers, which provide the ability to shift an entire data center’s workload to other facilities.

In a March interview, Urs Holzle, Google’s Senior Vice President of Operations, said the company typically uses manual tools to manage data center level outages and downtime.  “Teams regularly practice failing out of or routing around specific data centers as part of scheduled maintenance,” he said. “Sometimes we need to build new tools when new classes of problems happen.”

Redirecting Workloads Instantly
At last month’s Structure 09 conference, Google’s Vijay Gill hinted that the company has developed automated tools to manage data center heat loads and quickly redistribute workloads during thermal events (a topic covered by The Register).

“You have to have integration with everything right from the chillers down all the way to the CPU,” said Gill, Google’s Senior Manager of Engineering and Architecture. “Sometimes, there’s a temperature excursion, and you might want to do a quick load-shedding to prevent a temperature excursion because, hey, you have a data center with no chillers. You want to move some load off. You want to cut some CPUs and some of the processes in RAM.”

Gill was asked if this was a technology Google is using today. “I could not possibly comment on that,” Gill replied.

Look Ma: No Chillers!
But Google engineers had already disclosed the existence of the chiller-less Belgium data center at the Google Data Center Efficiency Summit in April in Mountain View, Calif. At the event, we asked specifically: are there chillers on-site that are rarely used, or no chillers at all?

The answer: no chillers at all. The facility will rely entirely on free cooling, and redirect workload on days when it’s too hot to operate. This approach makes local weather an issue in network management, although advanced forecasting can help Google anticipate days when it may need to divert work from the Belgium facility.

Nonetheless, even Google is periodically challenged by rerouting entire data centers, as seen in a February Gmail outage when a data center was overloaded while shifting workloads. Traffic redirection was also an issue in a brief outage in May.

An Enabler for “Follow the Moon”?
The ability to seamlessly shift workloads between data centers also creates intriguing long-term energy management possibilities, including a “follow the moon” strategy which takes advantage of lower costs for power and cooling during overnight hours. In this scenario, virtualized workloads are shifted across data centers in different time zones to capture savings from off-peak utility rates.

This approach has been discussed by cloud technologists Geva Perry and James Urquhart as a strategy for cloud computing providers with global data networks, who could offer a “follow-the-moon” service to enterprise customers who would normally build data centers where power is cheap. But this approach could also produce energy savings for a single company with a global network – someone like Google.