SolidNode

Digital Realty has agreed to purchase three sites from Sentinel Data Centers, including this facility in Needham, Mass.

Digital Realty Trust (DLR) said today that it plans to acquire three data center properties in Massachusetts and Connecticut, paying $375 million for 550,000 square feet of buildings operated by Sentinel Data Centers. The deal expands Digital Realty’s footprint in the Boston market, where its three existing properties are fully leased.

Digital Realty also said it may sell up to $400 million in common stock through equity distribution agreements with Citigroup, Merrill Lynch, and Credit Suisse. The shares will be sold under a shelf registration agreement filed with the SEC in May 2009. Under a shelf registration, a company may sell securities in one or more separate offerings with the size, price and terms to be determined at the time of sale.

The additional funds will help Digital Realty manage its borrowings as it continue to acquire leased data center properties. The company reported in an SEC filing that it entered into definitive purchase agreements on Dec. 24 to acquire three properties:

• 55 Middlesex Turnpike in Bedford, Massachusetts.
• A 100% condominium interest that represents 87.5% of the square footage of 128 First Avenue, Needham, Massachusetts, a 200,000 square foot facility initially built by Level 3 Communications.
• 60-80 Merritt Boulevard, Trumbull, Connecticut.

The properties have been renovated to “modern data center standards” within the last four years, the company said. The buildings have 35 tenants, with leases totaling more than 306,000 sauqre feet, which Digital Realty says represents about 98 percent of the leasable space. Those leases are currently generating annualized rent of about $40 million. Digital Realty said it plans to fund the acquisition from its revolving credit facility, and expects the deal to close by Jan. 15.

In October Digital Realty signaled its intent to buy additional occupied data centers as income properties, which generate revenue through rent from existing tenants.

Original post:
Digital Realty Pays $375M for New England Sites

Popularity: unranked [?]

Colocation provider Peer 1 said a technician was injured during an incident Wednesday night that knocked out power at its data center at 151 Front Street in Toronto. The company said the injuries “appear to be non-life threatening.”

The technician was injured during scheduled maintenance on the uninterruptible power supply (UPS) system in Peer 1’s fourth-floor data center at 151 Front, the largest carrier hotel and data center hub in the Toronto market.

“During our scheduled maintenance on the UPS in Suite 402, UPS shorted to ground, fire alarm was tripped, power was shutdown, and the technician was injured,” Peer 1Network Operations Center Manager Jason Mosher reported on the company’s customer forum. “We are doing everything we can to get power restored quickly. The technician is currently en-route to the hospital and we have all staff dispatched to assist.”

The incident occurred at about 7:20 p.m. Eastern time. All staff were subsequently cleared from the building while the fire and police investigated, but were cleared to return by 8:45 p.m. Power was restored at approximately 9:30 p.m., and Peer 1 said it had additional staff on hand to provide support for customer site restoration.

Original post:
Technician Injured in Peer 1 Power Outage

Popularity: 10% [?]

There’s a discussion today at Slashdot about how many servers or users an admin can manage. The question seeks to establish whether a department in which sysadmins each manage 900 user machines is understaffed. As in any discussion of IT or data centers, practices vary widely. But here are a couple of relevant factoids from our travels tracking the data center industry.

Jeff Rothschild, the vice president of technology at Facebook, said in a recent presentation that Facebook has 230 engineers supporting data for more than 300 million users. He says Facebook seeks to maintain a ratio of one engineer for 1 million or more users. Facebook is vague about exactly how many servers it has, saying it’s “more than 30,000.” But 30,000 servers and 230 engineers works out to a ratio of about 130 servers per admin.

Microsoft says it has automated its data center operations to the point where its admins can each manage between 1,000 and 2,000 servers. That matters, as the company may pack more than 300,000 servers into its new container data center in Chicago. It expects to support that facility with about 30 employees, including admins and facility maintenance staff.

Original post:
How Many Servers Can One Admin Manage?

Popularity: unranked [?]

Mark Bramfitt, who has been the electric utility industry’s most visible ambassador to the data center sector, announced in November that he will be leaving his post at PG&E. Bramfitt has been a familiar face at industry conferences as PG&E has rolled out a series of incentive programs to encourage data center operators to conserve energy by using more efficient technology and products.

What’s ahead for Bramiftt? And what does his departure signal about the future of utility-level incentives for data centers? Jay Fry at Data Center Dialog has posted a two-part interview with Bramfitt that provides some additional information.

Bramfitt will be doing consulting focused on “better engagement between the utility and IT industries,” he says in the interview. That will include “driving nationwide activities that will hopefully yield big results.”

He also reflects on the challenges in helping data centers and utility providers work together on improved energy efficiency. “What has slowed us down, I think, is that the IT industry and IT managers had essentially no experience with utility efficiency programs three years ago,” he says. “It simply has taken us far longer than we anticipated to get the utility partnership message out there to the IT community.”

In Part 2 of the interview, Bramfitt discusses the Utility IT Energy Efficiency Coalition, a consortium organized by PG&E that now includes 50 power companies. He sees an opportunity for the group to be more effective in bridging disconnects between the utility industry and data center operators.

“I think it’s time for the coalition, in whatever form it takes in the future, to expand to offering the IT industry a view of what utilities are offering programs and how to engage with them, as well as a place for IT and other companies to list their competencies,” he said. “I’ll be frank, though, in saying that I don’t know whether PG&E will continue to lead this effort, or if we need to think about another way to accomplish this work.”

Go to Data Center Dialog to read the entire interview.

Original post:
Mark Bramfitt on Data Centers and Utilities

Popularity: unranked [?]

Ever want to see inside the world’s most powerful data centers? We’ve provided a look inside many of the world’s most interesting facilities here at Data Center Knowledge. Here are our picks for five of the coolest video tours of major data centers, along with a list of links to 10 other worthwhile video tours.

These videos provide walk-throughs of data centers operated by Google, Equinix, Terremark, Sun Microsystems, Intel, IBM, Switch Communciations, Wordpress.com, Nvidia, Emerson Network Power, Telecity, the CLUMEQ supercomputing center and Bahnhof.

For many more data center videos, check out our DCK video archive and the Data Center Videos channel on YouTube.

Original post:
Coolest Data Center Video Tours

Popularity: unranked [?]

The biggest deal in the data center industry for 2009 was the news that Equinix (EQIX) would acquire Switch and Data in a $689 million transaction. As the year draws to a close, we’re learning more about the behind-the-scenes maneuvering that led to the deal, in which Switch and Data (SDXC) spent months weighing competing bids from Equinix and an international provider.

The negotiations with the rival suitors, which began in earnest in April and concluded in late September, are detailed in an SEC filing by Equinix. The existence of a rival suitor for Switch and Data was first noted by Paolo Gorgo in an analysis at Seeking Alpha last week.

In January Switch and Data was approached by “a privately-held firm in its industry based outside of the United States” to discuss a possible business combination. This international company proposed a merger of equals with Switch and Data, which emphasized that it was not for sale but began discussions in March.

Equinix contacted Switch and Data in April to express interest, kicking off five months of talks. Equinix dropped out of the talks in mid May, but resumed discussions in early July. On Sept. 26, the Switch and Data board determined that the Equinix bid – which represented a 32 percent premium to its stock price – was the best outcome for its shareholders.

Who was the mystery suitor? Paolo notes that the description fits Interxion, which already had a joint marketing agreement with Switch and Data. At the time, we noted that the partnership “appears to be a competitive response to the growth of Equinix, which acquired IXEurope last fall, providing it with facilities in major Internet markets on both sides of the Atlantic.”

The description could also fit KDDI, the Japanese company that owns colocation and peering providers Telehouse America and Telehouse Europe. KDDI has been actively expanding its Telehouse network in Europe and Asia, but has not announced plans for expansion in the U.S. beyond its three existing facilities.

Averting A Global Competitor
By stepping up to outbid the international firm, Paolo notes that Equinix was able to “avoid the creation of a global competitor in the sector, and stopped a foreign entity from achieving a US listing (without going through an IPO) and a strategic footprint in the most important market for the colocation sector.”

The acquisition was also driven by the economics of data center development, allowing Equinix to enter new geographics and add capacity in key existing markets for less than it would have cost to build new facilities in those locations.

The Equinix-Switch and Data deal combines two of the largest players in the market for colocation and interconnection services, and gives Equinix a presence in 16 new markets across North America, including Atlanta, Denver, Miami, Seattle and Toronto.

Equinix will integrate Switch and Data’s data center business and operations, including the company’s 34 data centers in 22 markets in the U.S. and Canada. The acquisition will add more than 1 million gross square feet of data center capacity, bringing Equinix’s total global footprint to more than 6 million square feet across 79 data centers in 34 markets in North America, Europe and Asia-Pacific.

Vote Set For Jan. 29
The deal is currently undergoing regulatory review, and Switch and Data shareholders are scheduled to vote on the Equinix proposal at a Jan. 29 shareholders meeting.

The deal may have additional ripples as competitors weigh their strategic options going forward. The Sunday Times of London reports that Interxion may be planning an IPO in 2010. The paper says the company has been in discussions with investment banks about a public offering on the NASDAQ market.

It’s important to note that this is not the first time UK media have predicted an Interxion offering. Exactly two years ago the Guardian wrote that Interxion was discussing an IPO with investment banks. Instead, the company remained privately held and raised $190 million to continue its aggressive expansion of its data center network, which now spans 26 facilities in 11 countries.

Original post:
Switch and Data Had Another Suitor

Popularity: unranked [?]

With all of the attention IBM, HP and now Cisco get in the server market, the number three player in the server business sometimes gets overlooked. But Dell (DELL) has been quite active this year in both the enterprise server market and the evolving market for cloud-optimized servers and containers.

ServerWatch has a nice article about Dell, its market position, and an in-depth look at the 11th generation of servers.  The article details the entire PowerEdge line of servers, comparing target deployments, processors, operating systems, and pricing.

The Register discusses Dell’s “other server business” earlier this month by digging into the OEM Solutions Group at Dell. The Data Center Solutions (DCS) group within Dell is responsible for custom solutions sold to hyperscale data centers such as Yahoo, Amazon and Microsoft. The OEM Solutions group is apparently twice the size of DCS, according to Dell’s Rich Froehlich. Citing OEM deals such as medical equipment, kiosks, embedded controllers and various appliances, the OEM group has 1,300 customers in over 40 countries and leverages 6,000 engineers to work on design and six world-wide factories to deliver the finished goods. 

Dell won the Google search appliance deal and there was a “container system in the works” reported in May 2008.  That container data center was later confirmed to be used in the Microsoft Chicago container farm.

Tough server market
In tough economic times and and an even tougher server market, Dell has held its own.  Gartner reportedin September 2009 that second quarter worldwide server shipments dropped 28 percent year-on-year, while revenue fell 29.4 percentage points year-on-year.  Gartner lists IBM with a 32.5 percent share, HP at 28.9 percent and Dell with 13.3 percent of second quarter server revenue estimates.

For server vendor shipment estimates Gartner puts Dell second, with 23.9 percent, behind HP’s 31 percent. Data from IDC shows third quarter server sales taking a steep decline.  IDC market share numbers put Dell in third place as well with a 13.5% market share. Sun was listed in fourth but it was also noted that both IBM and Dell were taking market share away due to the prolonged delay in its takeover by Oracle.

Converged data center offering
Dell spent much of 2009 building partnerships and growing all facets of its business.  Partnerships were inked with Brocade and Juniper and in September Dell acquired Perot Systems for $3.9 billion. Recently Ross Perot Jr. joined the Dell Board of Directors. 

Dell also maintains 16 spots on the top500 list for November 2009.  Dell maintains a partnership with Cisco, but also does not see the value in their Unified Computing System.  Dell’s Praveen Asthana was quoted in The Register as saying that “there’s a difference between a proprietary stack and a reference architecture.  Our approach is to offer reference architectures, but we don’t restrict you to buying everything from Dell.”

Social networking saves the day
Slate’s The Big Money site had an article last week explaining how “How Dell Got Out of Hell”.  Dell has made great strides from former customer relations and support issues and now has been very active in social networking. The article explains how Dell has recorded $6.5 million of sales using Twitter, using more than 100 employees tweeting through 35 Twitter feeds. The LinkedIn page for a company is always interesting to read.  Dell’s shows that career paths for employees before, came from HP and IBM, and after, went to Microsoft and HP.  Dell employees are most connected to Equallogic, Microsoft, CSC and Oracle.

The server side of the business will have a little bit of that spotlight taken away though, as Dell is rumored to unveil an Android based tablet at CES and last month confirmed its smart phone plans.

Original post:
Busy Year For Dell in the Data Center

Popularity: unranked [?]

Duncan Campbell is vice president of marketing for Converged Infrastructure in the Enterprise Storage, Server and Networking Worldwide Organization at HP.

DUNCAN CAMPBELL
HP

As data center managers prepare for the New Year, they should consider “reinventing” their infrastructure the same way many people reinvent themselves on Jan. 1 by beginning an exercise program, quitting smoking, or sporting a new hairstyle. Particularly as 2009 brought a whirlwind of new trends and the products to support them, this is a good time to take a hard look in the mirror, re-assess your technology infrastructure, and decide on a path to data center transformation.

In advance of 2010, what follows is a list of New Year’s Resolutions to ensure a seamless transition into the New Year:

1. Become More Flexible: With more than 70% of IT budgets still dedicated to operations and maintenance, this is the year that data center managers should look to finally break down IT silos for a more flexible technology infrastructure that can respond instantly to business demands. New technologies and trends in computing that foster data center convergence will help make this longtime vision a data center reality in 2010.

• Convergence integrates existing silos of compute, storage, network and facility resources with unified management to deliver a virtualized, highly-automated technology environment. With pools of shared-services that can be leveraged on-the-fly, data center managers can improve the flexibility of their environments, while speeding time to application value.

2. Go More Green: The No. 1 problem facing today’s data centers is hitting limits of power and cooling capacity. As it becomes imperative to better control costs, a blueprint of the data center that provides a more intelligent view of the entire IT environment –including the facilities – will help data center managers reclaim stranded power and cooling capacity. To enable new levels of energy and cost efficiencies, facilities must share a common, comprehensive view of data center power and cooling. This requires a significant shift from the way most data centers are currently managed, to one in which facilities and IT managers join forces to tackle energy issues datacenter-wide. Instead of Green IT think of it as IT for “the green” that helps the pocket book and the environment!

• With intelligent monitoring systems, energy use can be viewed, monitored and controlled in real-time, while users can establish policies for energy use to ensure compliance with internal, environmental or government regulations and goals. New power monitoring systems can extend across IT systems to include facilities as well, considering power consumption as important as systems performance in our efforts to green the data center.

3. Make Over My Infrastructure: With server maintenance contracts set to expire after approximately three years and many customers delaying server hardware refreshes since the economy slowed, 2010 presents a good opportunity for companies to refresh their server farm. According to Gartner’s Top 10 in 2010 forecast discussed at Gartner Symposium, the cost of operating a server in three years’ time (the typical lifecycle of a server) will eventually surpass the original purchase price of the server, pointing to the need for more energy efficient computing products.

• Customers should take advantage of the latest technologies available in servers – like power capping, embedded power management, and sensors – to ensure the power being consumed is only what’s required for the work being performed. The combination of these capabilities result in slashing server energy costs by 50 percent.

4. Be Open-Minded: With standards-based networking products, organizations will avoid locking their infrastructure into restrictive proprietary networking solutions that entail arduous upgrades and extremely high maintenance costs. Most importantly, proprietary networking products lack the key components necessary for achieving optimal centralized network management.

• Now more than ever, it’s becoming harder to justify the expense of proprietary software. Networking infrastructure is difficult to build and even more difficult to operate as the network grows in size and complexity. Companies should take advantage of open standards-based networking solutions that provide freedom now and into the future – by leveraging the right balance of applications and hardware, companies can optimize business results while breaking free from restrictive proprietary technology.

5. Think Outside the Box: In order for customers to reap the full potential of a virtualized infrastructure, business need to virtualize their data dater from end-to-end. By converging servers, storage, and networking resources into a single pool, customer can increase performance, scalability and capacity, while simplifying management.

• Storage plays a critical role in infrastructure convergence by allowing customers to virtualize stored data and create a unified virtual resource pool that is instantly accessible to support changing business needs. To maximize data center efficiency as well as reduce costs, customers need storage solutions that accommodate data growth, are aligned to business applications and are built on open standards.

Original post:
New Year’s Resolutions for Data Center Managers

Popularity: unranked [?]

Some of the web’s most prominent sites experienced downtime and sluggishness Wednesday night due to problems with DNS services. The issues were most pronounced at UltraDNS, which reported that its performance problems were caused by an electronic attack.  

The list of sites experiencing problems included Salesforce.com, Amazon Web Services and Walmart.com. DNS is short for the domain name system, which serves as a roadmap allowing users to find web sites. Domain registries like VeriSign provide centralized web lookups through a network of data centers, while commercial DNS service providers like UltraDNS offer additional tools to manage traffic.

UltraDNS told CNet that it was hit by a distributed denial of service attack (DDoS) targeting its west coast infrastructure at data centers in San Jose and Palo Alto. A DDoS attack targets a site or provider with large volumes of traffic in an attempt to overwhelm its ability to serve content. 

That logjam at UltraDNS caused ripples across the Internet, causing uptime  problems for several major servide providers. 

“We can confirm some customers in the West Coast are experiencing issues with resolving DNS,” Amazon Web Services reported. The issue was resolved by 6:40 p.m. and affected Amazon’s EC2 compute service, S3 storage and CloudFront content delivery network,  Pacific time. 

“Due to a global DNS outage, all services at salesforce.com were affected,” Salesforce reported on its status dashboard. ”We are closely working with our providers to get more details and to isolate the nature of the problem.”

The outages marked the third time this year that DDoS attacks on DNS providers had disrupted service for customers:

• Hundreds of thousands of web sites were slowed for hours on Jan. 23 when Network Solutions’ DNS servers were hit by a DDoS. The company described the incident as a “very large scale” attack.
• UltraDNS blamed an April 1 outageon a DDoS, but said the issue affected a small subset of its customers.

Successful attacks on DNS providers are not unprecedented, but these services are designed to be resilient. VeriSign has deployed new infrastructure in a $100 million upgrade at multiple data centers, witha  goal of expanding its capacity to handle more than 4 trillion DNS requests per day by 2010.

Original post:
DNS Issues Cause Downtime for Major Sites

Popularity: unranked [?]

North Carolina has been a hot market for data center development in the past two years, as the combination of affordable power and aggressive tax incentives have lured major new projects from Google and Apple.

T5 Partners would like to help accelerate the trend. T5 is a data center development company led by former data center specialists with The Staubach Company. After years of performing data center site selection for enterprise clients like T-Mobile and E-Trade, the T5 team decided to focus on the other side of the business.

The company is now developing existing structures and sites that can be quickly converted into major data centers, and is focusing its initial efforts on marketing two properties near Charlotte for data center development.  

NC Wants Data Centers
“Choosing North Carolina first is no accident,” said Peter Marin, president of T5 Partners. “It’s based on the cost of power, incentives, low taxes and bringing the infrastructure together. The incentives are aggressive in North Carolina on purpose. They want to attract data centers.”

T5 doesn’t discuss its current clients, but the company is known to be working closely with officials in Catwaba County, where Apple is building a new $1 billion data center campus. State and local incentive packages are expected to result in about $46 million in rebates to Apple over the next 10 years.

T5 is focused on sites with existing structures and industrial-strength infrastructure in place, including adequate power and water supplies. The local utility for both its projects is Duke Energy, wich offers industrial power rates as low of 4 cents per kilowatt hour. T5’s target market is Fortune 500 companies who want to control their own data center, but have a tighter time frame that may not work with a ground-up “greenfield” construction project.    

“The option to fund and lease these facilities is very attractive,” said Jason Chartrand, executive vice president of T5 Partners. “Our buildings are basically ready to go and ready for tenant improvement.”

Sale/Leaseback As An Option
Chartrand noted that T5 is also working on arranging sale/leaseback deals, in which a property owner sells their building, while agreeing to continue to lease space in the building. The transaction generates cash for the former owner (now the tenant), and provides the new owner steady rent from the lease. 

For companies looking for a wholesale data center solution, T5 is working with Power Loft, which recently opened its first data center in Manassas, Virginia. “We’re able to address the speed-to-market issue with Power Loft,” said Marin. “We can get someone operational in 90 to 120 days.”

Both Power Loft and T5 Partners are backed by the real estate investment fund Iron Point Real Estate Partners, a real estate fund with offices in Washington, D.C. and Dallas.

Original post:
T5 Targets North Carolina for Development

Popularity: unranked [?]