Friday, September 6th, 2013

Time	Event
11:41a	Hot in Vegas: Switch Adds 19 Megawatts of SuperNAP Sales One of the huge data center power spines at the Switch SuperNAP campus in Las Vegas. Switch reports that it sold 19 megawatts of power to customers in August. (Photo: Switch) The data center market was plenty hot in Las Vegas in August. Colocation provider Switch said it sold 19 megawatts of  power capacity during the month, its highest one-month total ever. The majority of that capacity was for space in SuperNAP 8 , the company’s newest mega-data center in Las Vegas, which opened its doors earlier this year. Switch said the original SuperNAP, a 400,000 square foot facility built in 2009, is nearly filled. The 19 megawatts of power was sold to 23 customers, reflecting the scale of the new requirements companies are planning for SUPERNAP 8. Most colocation companies deal in customer requirements up to about 500 kilowatts, with larger requirements seeking out “wholesale” turn-key data center suites. Switch is among the colocation providers supporting larger power requirements, enabled by the company’s focus on containment systems to manage high power densities, allowing customers to house more IT capacity in a smaller footprint. SuperNAP 8 is part of a multi-phase expansion announced by Switch in 2011, which will include at least one more data center on its current Las Vegas campus. The company says the new facility provides capacity for Switch to house up to 20,000 cabinets on its campus, with capacity of up to 200 megawatts of power. Leveraging the SuperNAPs’ Collective Buying Power Switch says the strong sales are driven by a combination of location, massive power and big pipes, a recipe that has helped the SuperNAPs address scalability and density challenges for customers. A key asset is a purchasing co-op known as the Telecom Combined Operating Retail Ecosystem (CORE), which harnesses the combined buying power of its community of high-density customers to gain discounted bulk rates from carriers. The company’s Las Vegas campus has also benefited from a concentration of cloud computing providers, including HP, Joyent, ProfitBricks, CloudSigma, VMware and many others. Switch’s 600 customers in its Las Vegas data centers include more than 80 companies offering either connectivity or cloud services. “Most of the industry knows that Switch designs and builds the world’s best data centers, but our unmatched differentiators are Switch’s independent co-ops to purchase connectivity, cloud and content for our clients,” said Switch Executive Vice President of Colocation Missy Young. “Once enterprise clients start to work with those purchasing verticals they quickly realize that no other ecosystem in the world can bring about a total technology return on investment like the SUPERNAPS.” Infrastructure Innovation Continues Young said Switch has also differentiated itself with continuous innovation in its infrastructure, driven by Switch CEO and Founder Rob Roy. SuperNAP 8 includes a number of refinements to Switch’s custom data center designs, including a double-reinforced roof, an updated rack containment system that incorporates thermal storage, and a refinement to its custom cooling units to add an on-board backup power system. SuperNAP 8 features Switch SHIELD, a redundant data center roofing system that offers two steel roof decks, each rated to survive winds of up to 200 MPH. The two roof decks are located nine feet apart and are attached to the concrete and steel shell of the facility and contain no roof penetrations. Switch has also introduced Rotofly, an expanded version of its multi-mode HVAC unit that incorporates a flywheel to provide emergency backup power to keep the cooling system operating during a power loss. A flywheel, also sometimes known as a rotary UPS, is a spinning cylinder that generates power from kinetic energy, continuing to spin when grid power is interrupted. The flywheels are only being used to support the cooling units, while the servers and other mission-critical equipment will continue to be supported by battery UPS units and backup generators. “It is our mission at Switch to develop smarter systems and think beyond tomorrow for the benefit of our clients,” said Young. “We take pride in providing our clients with a unique ecosystem of scalable solutions and an exponential return on their investment.” A look at some of the server and cabling density found inside the Switch SuperNAP campus in Las Vegas. (Photo: Switch) (Comment on this)
12:30p	The Future is Converged: How to Overcome Infrastructure Integration Challenges Radhika Krishnan is vice president of product marketing and alliances of Nimble Storage, a provider of flash-optimized hybrid storage solutions. RADHIK KRISHNAN Nimble Storage No man is an island, said the famous English poet John Donne, and no storage product is an island either. Storage infrastructure will soon move to primarily pre-validated hardware and software systems, known as reference architectures. Industry analyst firms, like Wikibon, have predicted that over half of the entire infrastructure consumed will move towards pre-validated reference architectures, rather than independently procured applications, servers, hypervisors, networking and storage products. Analysts predict over half of the entire infrastructure consumed will move towards pre-validated reference architectures. In fact, the industry is already seeing this transition. There are two ways for enterprises to create a converged infrastructure: 1. The first is through the use of a single large vendor such as HP, Dell and IBM. These companies have repositioned their portfolio of products as converged infrastructure, which appeals to customers because they’re able to deal with a single provider to get an integrated solution for storage, networking, and compute. However, the concern many enterprises are voicing is that, while these companies have their sweet spots – e.g., Dell for servers – it’s not in their heritage to offer best-of-breed technologies across an entire network architecture. 2. The other solution is for enterprise IT departments to purchase individual best-of-breed servers and software that best fit particular needs and incorporate them into a single system. While these best of breed offerings provide a better solution than the one vendor approach previously mentioned it is still not an ideal approach. The challenge in these instances is that customers often lack the time or expertise to integrate best-of-breed servers, hypervisors, networking and storage. Instead, true reference architectures should address both problems delivering best-of-breed products while also relieving the customer from having to integrate best-of-breed products from multiple vendors. Industry Outlook to Assure Competition Preconfigured architectures from larger vendors are already appearing, but they don’t provide an ideal solution to all customers. In order to maintain a competitive market and keep the incumbents from dominating unchallenged, the market needs smaller vendors to start thinking about how to partner with other vendors, both large and small, to develop their own converged infrastructure. In fact, if they haven’t already started, they are already behind. As with anything else, not all converged infrastructures are created equal. When choosing a pre-validated reference architecture there are three characteristics that an IT buyer should keep in mind when determining which is the best fit: First, the solution must be economically viable. Resistance to change is natural, and this is especially true in IT. It might sound obvious, but the solution must bring economic benefits for it to gain acceptance from the C-suite. This benefit can come from higher performance, lower CAPEX or OPEX, improved business continuity or operational simplicity to the company. More importantly, these benefits have to be provable. Second, the solution must be truly “best-of-breed.” Be cautious of buying pre-validated and pre-configured architectures. Don’t assume that, because you’ve heard of one vendor or a vendor has a good reputation, the entire architecture will be best-of-breed. You must become familiar with each technology in the architecture, but there’s no burden on you to integrate them! After all, the point of a pre-validated architecture is that the component pieces come pre-integrated. Still, your best protection is to research the individual vendors. Lastly, leverage the channel, either via value added resellers (VARs) or managed service providers (MSPs). This is important for support after implementation. Utilizing the channel will provide a single point of contact for any issues that may arise for any part of the solution providing more efficient support as needed. As time goes on, converged infrastructure will continue to become more prevalent in the data center. If executed properly, it can provide many benefits for understaffed and overstressed IT departments. Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library. (Comment on this)
2:11p	Penguin Demos Open Compute Microserver With New Intel Atom C2000 SuperMicro has introduced a new 6U microserver using the new Intel Atom C2000 processor. (Photo: SuperMicro) With the Wednesday launch of Intel’s Atom C2000 processor, partners Penguin Computing and Super Micro have launched products that feature the new low-power chip. Penguin Computing OCP Microserver Pengiun Computing demonstrated a new microserver platform based on the Intel Atom C2000 processor product family. It is one of the first implementations that is built according to the Open Compute Project’s microserver card specification, and can house the 64-bit Atom C2000, with power envelopes as low as 6 watts. With the C2000 at its core the microserver platform features a modular architecture that allows for  right-sizing compute capacity by populating each chassis with the right number of microserver cards required to accommodate a specific workload. The microserver cards as well as the microserver system are compliant with specifications defined by the Open Compute Project. “Penguin Computing is at the forefront of delivering cutting edge server and storage solutions for the Open Data Center and our customers are focused on delivering the maximum compute and storage density within existing power and cooling constraints,” said Charles Wuischpard, CEO of Penguin Computing.“Due to its ultra-low power envelope, high density, expanded memory and 64-bit support for the x86 instruction set we’ve added designs based on the Intel Atom processor C2000 to our product roadmap that will be fundamental to our microserver-based solutions.” Super Micro MicroBlade Super Micro (SMCI)  debuted a new 6U MicroBlade product – a microserver featuring 112 ultra low power 8-core Intel Atom C2000 processor-based servers in 28 hot-swap micro blades. The modular architecture maximizes rack space with full-featured power-conserving servers in easily serviceable front access hot-swap Blade trays. Compute and storage are integrated in the individual blades and shared networking, power and cooling are located at the rear of the system. “Supermicro continues a tradition of innovation and green computing with the debut of our new extreme-density, high-efficiency MicroBlade,” said Charles Liang, President and CEO of Supermicro. “MicroBlade is engineered with our most advanced power saving, modular architecture designs and incorporates Intel’s latest low-power Atom C2000 processors to dramatically increase compute density and maximize performance per watt, per dollar, per square foot. With 112 full featured energy efficient compute nodes per 6U, MicroBlade brings a new cost-effective, environment friendly microserver option to the table for rapidly expanding data centers and cloud service providers.” (Comment on this)
3:22p	Oracle Launches Government Cloud Solution With the specific target of government entities needing a secure cloud, Oracle (ORCL) announced the availability of the Oracle Government Cloud, a comprehensive suite of cloud applications and solutions. Oracle is currently seeking the required authorizations designed to provide government agencies with this solution. Solutions within the offering include Oracle Service Cloud, Oracle RightNow Policy Automation and Oracle Learn Cloud. The will assist agencies in streamlining a breadth of business processes, from financial and human resources management to customer service and project management. Oracle Government Cloud will enable agencies to leverage multiple service options – with Software as a Service (SaaS) offerings available today and Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) . “We are very excited to meet our public sector customers’ demands for a highly secure, robust suite of cloud solutions built for government. The maturity and transparency of the standards-based Oracle Government Cloud will enable agencies to confidently move mission-critical government applications to the cloud without compromising security, performance, or reliability. We are committed to helping agencies enhance operational efficiency through the cloud,” said Mark Johnson, senior vice president, Oracle Public Sector. As an Oracle Cloud solution, the complete portfolio of public, private and hybrid cloud offerings is owned, managed and certified by Oracle. Th solutions gives customers performance, scalability and security needed to run mission-critical, government-specific solutions. Oracle OpenWorld will be in San Francisco later this month. (Comment on this)
3:23p	Friday Funny: Extending More Power It’s Friday! And you know how we like to end the work week with a little humor. Our Data Center Knowledge cartoon contest is a perfect way to have a few laughs on a Friday. Diane Alber, our data center cartoonist, writes, “Running out of power or power strips, or patch cords can be a pretty common problem. . .Let’s see how Kip and Gary fix this one. . .” Submit your caption suggestions below. Also, big congratulations to our reader JL, who submitted, “The new methane generators have arrived.” for our cartoon, How Remote Do You Go?. New to the caption contest? Here’s how it works: We provide the cartoon and you, our readers, submit the captions. We then choose finalists and the readers vote for their favorite funniest suggestion. The winner receives a hard copy print, with his or her caption included in the cartoon! Click to enlarge graphic. For the previous cartoons on DCK, see our Humor Channel. Please visit Diane’s website Kip and Gary for more of her data center humor. (Comment on this)

<< Previous Day

2013/09/06 [Calendar]

Next Day >>