SolidNode

One of the exhibitors at last week’s Gartner Data Center Conference was Panduit Corp., whose booth featured some of the company’s gear to support Cisco’s Unified Computing System (UCS). Panduit’s Vineeth Ram provided us with an overview of the company and its Unified Physical Infrastructure solutions, while CTO Jack Tison provided a look at how Panduit’s products work with Cisco UCS. This video runs about 8 minutes.

For additional video, check out our DCK video archive and the Data Center Videos channel on YouTube.

Original post:
Video: Panduit Partners With Cisco on UCS

Popularity: 6% [?]

Here’s a roundup of some of some of this week’s headlines from the data center and hosting industry:

• Private clouds from Cisco and Savvis.  Savvis (SVVS) and Cisco (CSCO) announced an expanded relationship Monday that will focus on the development of private clouds for the enterprise. Formerly dubbed Project Spirit, the Savvis Symphony private cloud platform will integrate the Cisco Unified Computing System (UCS). This new platform will power the industry’s first enterprise-class Virtual Private Data Center (VPDC) with multi-tiered security and quality of service capabilities.  Savvis learned the benefits of the Cisco UCS solution early as they worked with Cisco to develop a reference architecture that will enable enterprise customers to take advantage of Infrastructure-as-a-Service (IaaS) benefits available in Savvis Symphony.
• Axxana Enterprise Data Recorder to be available through EMC Select. Data protection startup Axxana announced that its Phoenix System RP Enterprise Data Recorder will be available in January 2010 through EMC Select. The Phoenix System RP is a black box solution that was developed in cooperation with EMC’s RecoverPoint and delivers asynchronous data protection over any distance. ”Axxana’s Phoenix System will allow customers using RecoverPoint and CLARiiON networked storage to utilize replication over any distance, while maintaining the ability to recover data without data loss,” said Matt Mainstruck, manager of EMC SelectLast month Axxana secured $9 million in series B funding, led by Israel based Carmel Ventures.
• Enomaly launches cloud service provider edition. Cloud computing provider Enomaly announced the launch of 10 new global cloud computing service providers who have standardized and are powered by Enomaly’s Elastic Computing Platform (ECP), Service Provider Edition.  Called a “cloud in a box” solution for service providers, the ECP service provider edition is designed for carriers and hosting providers looking to build a line of business offering Infrastructure-on-Demand or Infrastructure-as-a-Service (IaaS) to customers.  The platform delivers a self-service dashboard and web-based API to manage virtual servers in whatever quantity needed.  Platform features include a multi-tenant, carrier class cloud platform with support for KVM, Xen and VMware hypervisors, detailed resource metering and accounting, tiered classes of service, a hard quota system and easy integration with back-office provisioning and billing systems.
• ColoHouse completes SAS70 audit. Miami data center and colocation provider announced completion of a SAS70 audit.  The audit focused on control environment, risk assessment, monitoring, control activities and information and communication systems. ”Obtaining our SAS 70 certification allows ColoHouse customers to rest easy knowing that we are following through on our commitment to provide the very best in Miami Data Center and Miami Colocation services, with a 3rd party validation to ensure our customers ultra secure access with a 100% uptime guarantee,” said Chief Operating Officer Alan P. Sabourin.

Original post:
Roundup: Savvis, Axxana, Enomaly, ColoHouse

Popularity: 6% [?]

Here’s a roundup of some of some of this week’s headlines from the data center and hosting industry:

• SoftLayer continues data center expansion. SoftLayer Technologies announced the opening of three new data center pods in the Dallas, Seattle and northern Virginia data centers.  The new pods add capacity for 20,000 additional servers, bringing total capacity to more than 45,000 servers. “These three new pods meet the customer demand increases that we expect in the very near future,” said Lance Crosby, CEO of SoftLayer. “And they are only preliminary measures in our growth strategy for 2010. We have some big plans which we can’t wait to share with everyone.”  SoftLayer has standardized on the pod architecture for data center design, allowing them to optimize space, power, network, personnel and internal infrastructure. The company recently announced that it was on track to report more than $80 million in revenue for 2009 and raised $20 million to fund the continued growth of the company.
• Power Loft opens Virginia Data Center: Power Loft LLC announced the substantial completion of their first data center, Power Loft @ Innovation. Located in Prince William County, Virginia, this 225,000 square foot facility has signed an international IT technology outsourcing company as its anchor tenant, and was recently awarded the first Northern Virginia Technology Council’s Green Award. “Power Loft is in the forefront of creating energy efficient data center space,” said Bobbie Kilberg, President & CEO of the Northern Virginia Technology Council (NVTC).  “Having our company singled out to receive the NVTC Green Award, turned four years of hard work into a very unexpected night of celebration for us all,” said Jim Coakley, CEO of Power Loft LLC. “We are very
  proud to be so honored and we commend the NVTC for elevating the visibility of the many companies in Northern
  Virginia who are making an increasingly positive impact on our environment.”
• Level 3 to support Clearwire’s 4G network. Level 3 Communications announced an expanded relationship with Clearwire Communications Tuesday to support their CLEAR 4G WiMax services.  The agreement provides Clearwire with network transport services as a part of their deployment of CLEAR WiMax services in major metropolitan markets across the United States. Level 3 will provide high speed connectivity to Chicago, Dallas, Philadelphia, Seattle, Washington D.C., Houston and the Bay area.   CLEAR 4G WiMax is a next generation mobile internet solution from Clearwire that claims to be 4 times faster than 3G.  Clearwire has been growing rapidly and in their third quarter 2009 results reported that 4G network coverage increased by 67% to over 10 million people.  They also recently had a $1.564 billion equity financing round. Is there a map for that? – you bet.

Original post:
Roundup: SoftLayer, Power Loft, Level 3

Popularity: 6% [?]

The Verari data center container housing the NASA Nebula cloud computing application arrives at Ames Research Center in Mountain View, Calif.

What do you get when you combine cloud computing and data center containers? You get NASA’s Nebula, the space agency’s new data powerhouse, which provides on-demand computing power for NASA researchers. Nebula was recently cited by federal CIO Vivek Kundra as an example of the government’s ability to “leverage the most innovative technologies.”

The Nebula application lives in a 40-foot container at the NASA Ames Research Center in Mountain View, Calif. The “data center in a box” was built inside a FOREST container from Verari Systems, which is filled with Cisco Systems’ Unified Computing System and servers from Silicon Mechanics.

Science Compute Power on Demand
Nebula is a self-service platform built from open source software that provides high capacity computing, storage, and network connectivity for NASA research. “Nebula has been designed to automatically increase the computing power and storage available to science- and data-oriented web applications as demand rises,” explains Chris C. Kemp, Chief Information Officer of NASA Ames Research Center.

“The containerized data center solution from Verari Systems and Cisco delivers the foundation for a next-generation cloud computing environment that is responsive to the needs of our developers so they can focus on mission success – without worrying about the capacity and availability of the computing infrastructure,” said Kemp. “his solution is transforming how we think about NASA’s future computing environment.”

Momentum for Containers
Data center containers allow for rapid expansion of IT infrastructure, and can provide excellent energy efficiency by offering more precise control of airflow within the container. Microsoft and Google have used containers as building blocks in  large data centers, while some enterprises, universities and research ;abs have used containers to add incremental compute capacity.

“Verari is simplifying data center deployment,” says Dan Gatti, senior vice president of Worldwide Market Operations, Verari Systems. “Our customers are able to meet their computing and storage requirements much more quickly and easily than ever before. The planning cycle for a data center has been cut from two years down to 120 days, on average. And our customers are able to recognize huge cost savings in both OpEx and CapEx spend.”

“Cisco and Verari Systems are delivering the data center of the future – today,” said Brad Boston, senior vice president of Cisco Global Government Solutions Group. “As NASA’s Nebula Cloud Computing Environment demonstrates, customers have a great deal of flexibility in how they integrate computing, storage, and networking capabilities with Cisco UCS to ensure a solution designed for mission success today and in the future.”

Federal CIO Vivek Kundra tours the NASA Nebula data center container during a September visit to Ames Research Center.

Original post:
NASA’s Nebula: The Cloud in a Container

Popularity: 2% [?]

Here’s a roundup of news announcements from the data center and hosting industry:

• Cisco’s Space Router: A “space-tolerant” router designed by Cisco flew into orbit this week aboard an Intelsat, Ltd. satellite. The Intelsat 14 satellite is a demonstration of Internet Routing in Space (IRIS) for the U.S. military. “We expect IRIS to connect the Internet with satellites in space for the warfighters who need seamless communication between ground-based networks and satellites used for communications,” said Kay Sears, president of Intelsat General. “Once the capabilities of IRIS are demonstrated, there could be a great deal of interest in this technology from a wide range of end users, both inside and outside of government.”
• Fujitsu opens annex of Tokyo data center. Fujitsu announced Monday the opening of the new annex of the Tatebayashi System Center in Gunma Prefecture outside Tokyo. The data center will power outsourcing services and next-generation cloud services business in Japan. The new annex also showcases the latest eco-friendly technologies as a part of “Fujitsu’s Green Policy Innovation” green IT project. The 247,000 square foot data center can accomodate 20 kVA per rack and employs solar power generation. The facility will initially contain a server farm configured with 1,000 servers and Fujitsu is looking to implement virtualization, autonomy, scalability, and usability, further advancing its Trusted-Service Platform for cloud services in Japan.

• Juniper products awarded MEF certification.  Juniper Networks announced Monday that its M series multiservice edge routers, T series core routers and MX series 3D universal edge routers have been awarded to Metro Ethernet Forum (MEF) 21 standards by Iometrix.  Juniper vice president of Operations Gert Jan Shenk said “MEF certification demonstrates that the Juniper routing portfolio enhances the operator experience and economics of networking for the service provider.”  Juniper unveiled its vision for a “new network” in October with new software, hardware and partnerships.
• Core NAP completes Type II SAS 70 audit. Austin Texas based Core NAP announced it has completed a Type II SAS 70 audit of its general controls supporting data center operations.  The report addresses the control environment, risk assessment activities, control activities, information and communication systems and monitoring activities.  “Core NAP’s management understands the ever-increasing importance of corporate governance, as well as the impact of the organization’s services on our clients’ internal controls,” said Kenneth Smith, president and CEO of Core NAP.

Original post:
Roundup: Cisco Space Router, Fujitsu, Juniper

Popularity: 12% [?]

NYSE Euronext says it has already pre-sold more than 20,000 square feet of colocation space in its new data center in northern New Jersey, reflecting strong interest in the exchange’s expansion into the data center services arena. The new facility, which opens next year, is the engine driving NYSE Euronext’s transformation into a technology platform for high-speed trading.

“We really see our data centers as the future of our market,” said Stanley Young, CEO of NYSE Technologies and co-global CIO of NYSE Euronext (NYX). “To show how vital this is, we have pre-sold space in that data center a year before our matching engines come online.”

Young discussed the NYSE’s new data centers and the role they play in its business during a panel as it hosted a recent energy efficiency summit with The Green Grid. Young said the 400,000 square foot data center in Mahwah will include 170,000 square feet of raised-floor data center space.

Large Footprint for Colocation
But NYSE Euronext will require just 20 percent of that space for its own infrastructure, Young said. The rest will be available for lease to trading firms seeking high-speed access to the servers – known as “matching engines” – that execute electronic trades.

In the increasingly data-driven environment on Wall Street, access to those servers has enormous value. “For our data center space I can get a multiple of three times what normal data center vendors can get for that space due to the proximity to the matching engines,” said Young. “You can see how valuable that real estate becomes.”

The Data Center as ‘Virtual Marketplace’
That has changed NYSE Euronext’s approach to its business. “We think of our data centers as the exchanges used to think of our trading floors years ago,” said Young. “What we’re seeing with our data centers is that we’re creating that virtual marketplace.

“We’ve actually changed our model,” he continued. “We’re moving from being a place where transactions occur to being what we call a fabric player. We allow industry participants to meet virtually and we get paid for that. That model is now driving all of our thinking.”

This evolution follows decades of market changes that have altered the primacy of the NYSE’s trading floor, once the iconic image of stock trading in the U.S. The emergence of computer-driven trading venues like the NASDAQ market has created a more diverse and competitive trading ecosystem.

Building A Platform Through Acquisitions
These forces have shaped NYSE Euronext’s adoption of technology to extend its reach as the rise of electronic trading shifted a growing percentage of financial trades to loe-latency platforms. In early 2006 the NYSE acquired electronic trading platform Archipelago Holdings to become a public company, and then expanded globally with a deal to buy EuroNext in 2007. 

The NYSE built its technology warchest with acquisitions of trading platform management specialists TransactTools (2006) and Wombat Financial (2008). It is currently in the process of consolidating these tools into a single Universal Trading Platform for the exchange and its partners.

NYSE Euronext sees this infrastructure – including the data center, network fabric and trading platform -as  its competitive edge in the new Wall Street. 

“In our business model we get paid by the transaction,” Young said. “(The number of) trades executed is a big revenue driver. If our market share is under pressure, being able to extend our platform becomes an important part of replacing our revenue from lost transactions.”

The Primacy of Power
As the activity shifts from the exchange post to the matching engine, those transactions are viewed in terms that matter most in the data center – the amount of power consumed.

“Our real business is power consumption and capacity of compute,” said Young. “That is our bread and butter. We sell space in our data centers by the kilowatt. We meter people’s cabinets. The more (trading) they do in those cabinets and the hotter they run, the more we earn. The cost of power is getting down to where people are making trading decisions based on it.”

That trend drove NYSE Euronext’s decision to get into the data center business, building its own facilities in northern New Jersey and London to anchor a next-generation trading operation. “We were running out of power, and had to look at our data centers, our trading floors for the future,” said Young. “We had to make some strategic decisions to build new data centers. It was critically important.”

Building From Scratch
At the time, NYSE Euronext operated out of data centers it leased from third parties. “When we looked around, there was nothing available that met the needs of our markets and business at all, so we decided to build from scratch,” said Young. “That gave us the opportunity to look at every single piece of green technology that was available in order to build these data centers as efficiently as we could.”

Young says billions of transactions every day will flow through the New Jersey facility, known internally as a “liquidity hub” rather than a data center. “You can understand why we build these data centers to 2N+1,” he said. “Everything is double backed up, at least.”

The facility will be able to support power densities of up to 14 kilowatts per cabinet. The company is using a modular design so it can deploy new space as needed.

The NYSE is focused on managing its power, but some popular efficiency measures are non-starters for the NYSE, including virtualization. “We can’t virtualize our trading engines because they’re running full blast all the time,” said Young. “What we’re starting to look into is putting shared memory access onto a matching engine.”

Looking at RDMA as Accelerator
NYSE has also worked with Cisco Systems on an Ethernet-based trading solution using Remote Direct Memory Access (RDMA) accelerators to reduce latency between compute nodes. RDMA allows data to move directly from the memory of one computer into that of another without involving either one’s operating system. Young says this technology  could result in latency of 10 to 15 microseconds when used over 100 gigabit Ethernet connections.

“Our world is a world where microseconds count,” said Young. “My users measure my abilty to serve them in tens of microseconds, and if they had the technology they would get down to nanoseconds.”

Original post:
NYSE Euronext’s Future: The Data Center

Popularity: 24% [?]

Power, cooling and space challenges are the norm for most data centers today. These challenges are a direct result of infrastructure sprawl and disparate technologies caused by cabling, adaptors and switch ports for networking and storage. While server virtualization has begun to address part of the issue, many problems remain and new challenges emerge. This webcast look as ways to solve some of these data center challenges using Fibre Channel over Ethernet.

The main application of Fibre Channel over Ethernet (FCoE) is in data center storage area networks (SANs). FCoE has particular application in data centers due to the cabling reduction it makes possible, as well as in server virtualization applications, which often require many physical I/O connections per server.

With FCoE, network (IP) and storage (SAN) data traffic can be consolidated using a single network switch. This consolidation can:

• reduce the number of network interface cards required to connect to disparate storage and IP networks;
• reduce the number of cables and switches;
• reduce power and cooling costs

Fibre Channel over Ethernet provides a homogeneous I/O such that all servers have access to both the Ethernet as well as Fibre Channel environments. This is key as workloads become more mobile and dynamic across the Data Center. Providing this incremental set of capabilities while optimizing the overall cable count, adaptors, switch ports, heat dissipation and power draw proves to be more and more compelling as servers, storage, and cables continue their otherwise unabated sprawl.

Many solutions are available today and IT organizations are realizing the benefits and efficiencies of products from Cisco, NetApp, and Panduit. This webcast will cover today’s challenges in the Data Center, perspectives on how to address them, and specific methods and products to mitigate infrastructure sprawl. Click here to register for this December 10, 2009 live webcast.

Original post:
Addressing FCoE in the Data Center

Popularity: 12% [?]

New colocation space is filling quickly when brought online, according to new data from the telecom research firm TeleGeography. Global colocation service providers surveyed by TeleGeography have added 1.66 million square feet (154,016 square meters) of new space since the beginning of 2008, many of which were more than 50 percent full by mid-year 2009.

“While ample capacity is available in aggregate, operators in some cities, such as London, San Francisco, and Los Angeles, reported fill rates of more than 80 percent” for their colocation footprints, TeleGeography said in announcing an update of its Colocation Database, one of the company’s paid services.

“Despite a global capital crunch, capacity growth has not stagnated, and operators with strong operational cash flow continue to build new sites,” it reports. The markets with the most new colocation supply have been Hybderabad, India (230,000 square feet), Washington/Northern Virginia (214,000 SF), Dallas (166,000 SF), Amsterdam (150,000 SF) and London (131,000 SF), according to TeleGeography.

The report also includes data on average power density per rack, which suggests that colocation providers are continuing to build ahead of the average power loads. “Colocation operators increasingly push boundaries in the installation of new high-density space (defined as space using at least 14 kilowatts of power per rack), but low-density spacecontinues to supply a majority of the market,” Telegeography writes. “The average colo site density among sites surveyed is about 3.8 kilowatts per rack, and major metro areas generally do not veer far from this average.”

Average density rates can be misleading when it comes to the potential business value of high-density space, however, as the average can include several high-density customers whose impact is broadly offset by customers with lower densities. In practice, the availability of space engineered for higher densities is crucial to capturing these deals, which can feature high-value customers.

The markets with the highest average densities were Chicago (5 kW average per cabinet) and Dallas (4.5 kW average).

Original post:
Report: New Colocation Space Fills Quickly

Popularity: unranked [?]

Here’s a roundup of news announcements from the data center and hosting industry:

• Force 10 demonstrates at Interop. Force 10 Networks partnered with AFORE Solutions and Sun Microsystems at INTEROP in New York last week to demonstrate data center virutalization products and cloud computing initiatives focused on driving greater network agility. The 3 vendors demonstrated virtualized resource allocation for cloud-oriented applications as well as the AFORE ASE3300 Virutalization Extension Platform.  The ASE3300 and Force 10 switch and router solutions combine to enable a multi-site, virtual data center enabling migration to cloud computing environments.
• Avocent upgrades data center management software. Announced at Interop last week, Avocent is upgrading its MergePoint Infrastructure Explorer to include several new management capabilities.  The company said these enhancements will provide a unique view into capacity planning, bringing additional return on investment and total cost of ownership benefits.  Avocent CTO Ben Grimes said that the software will allow “customers to know where their assets are, as well as plan for different ‘what if’ scenarios, and manage their data centers to reduce risk -  all while bringing improved ROI and total cost of ownership benefits to customers.”  New features include rack timeline and an enhanced change management and capacity search capabilities.

• Deutsche Telekom Buys Strato: Deutsche Telekom has acquired German web hosting provider Strato from Freenet, the companies said on Friday. Deutsche Telekom reportedly paid $409 million for Strato, which hosts 4 million domains, “This step boosts our position on the highly interesting market for hosting solutions,” said Niek Jan van Damme, Member of the Board of Management of Deutsche Telekom for fixed-network and mobile business in Germany. “Strato complements our activities in the hosting area perfectly and will make a positive contribution to net profit and free cash flow from the very first day of consolidation,” said van Damme.
• Social Networks: Facebook, Twitter, MySpace. Several news stories shed light on popular Web 2.0 sites last week.  Bloomberg estimates that common stock valuation for Facebook jumped 42% in the past four months to $9.5 billion.  The valuation comes after increased trading activity on stock service sites that allow current and former employees to sell shares of stock. Twitter COO Dick Costolo spoke at a conference last week and said that Twitter is getting more than $4 million a year in revenue from companies that use Twitter data.  Costolo also said that “you will see an advertising strategy from us in the very near future.” Cnet News reports that MySpace has aquired social-networking music site Imeem. Sources with knowledge of the deal say it is worth $8 million.
• Meanwhile, original web power-house company AOL has cut local headcount in northern Virginia by 2,400.

Original post:
Roundup: Force 10 Networks, Avocent, Facebook

Popularity: 32% [?]

The city of Prineville, Oregon is negotiating with a large, secretive company that wants to build a data center in its enterprise zone.

A “well-funded, well-known company” is negotiating to build a large data center in central Oregon, and the secrecy surrounding the negotiations has folks in the town of Prineville wondering who it might be. Officials in Prineville have been negotiating with Vitesse LLC, a company performing site selection for the unnamed end user that would build operate the data center, according to local media reports.

The site is several hours from an existing Google data center in The Dalles and a Boardman site where Amazon is said to be resuming construction on a major data center project. Like those projects, the process in Prineville has been cloaked in secrecy.

Google-Style Secrecy
“The only thing I could tell you is this is not unlike what the city of The Dalles went through when Google sited their data center in their community,” Prineville City Manager Steve Forester told the Bend Bulletin. “A very similar process. They had a code name for an LLC that did their preliminary work with the city and the county, and it turned out to be Google. And up in Moses Lake, Washington — where they have several of these things — same pattern.” Local economic development officials told the paper that a non-disclosure agreement prevents them from discussing the project.

Oregon business registration records indicate that Vitesse LLC was registered Oct 21 and shares a San Francisco address with the law firm Paul, Hastings, Janofksy & Walker. Attorneys with Paul, Hastings have data center site acquisition experience, including past engagements with large financial companies and Internet companies.

The proposed facility would be located near the Prineville Airport in an enterprise zone, which allows the city to waive property taxes for eligible projects. Tomorrow the Prineville City Council is scheduled to consider selling a 1-acre piece of property to Vitesse for $50,000, annex two adjacent properties to the city and approve a 15-year property tax exemption for the company that would operate the data center.

Oregon’s Lure for Data Centers
The project is the latest indicator of the growing appeal of the northwest as a destination for companies seeking the lowest operating costs for their data centers. The region’s abundant supply of affordable hydro power is a major factor in its appeal, as are tax incentives like the tax exemption being discussed in Prineville.

Who is the mystery user? The existing Google and Amazon projects in Oregon would seem to rule them out. But one possibility is Yahoo, which has publicly discussed the possibility of shifting some of its data center development to Oregon in the wake of a tax dispute with officials in Washington state, where the company built a large data center in Quincy.

Washington Repeals Tax Break
In late 2007 Washington State ruled that data centers aren’t manufacturers and were no longer covered by a state sales tax break for manufacturing enterprises, and thus must pay a 7.9 percent tax on data center construction and equipment. This prompted protests from Microsoft and Yahoo, who said they had relied upon the tax break in their decision to build facilities in Quincy.

In a letter to legislators, Yahoo co-founder David Filo said the withdrawal of the sales tax incentive “swings the decision strongly in favor of freezing construction in Washington, and building instead in Oregon (which has no sales tax), as some of our competitors are already doing.”

Microsoft subsequently migrated its Windows Azure cloud computing infrastructure from its data center in Quincy to another Microsoft facility in San Antonio.

Photo of Prineville, Oregon from wka via Flickr.

Original post:
Company X Plans Oregon Data Center

Popularity: 12% [?]