Oracle President Mark Hurd made it clear that Oracle has no intention of building mega-scale cloud data centers, stating that Oracle will “do what we have to do based on markets. We’ll have to put capabilities in place based on needs. It is all customer-centric. We don’t need data centers to be as large or exascale because the raw power of our machines has increased so much that it requires a smaller footprint, and we can house it in a smaller data center. Location is important, but it will all be customer demand drive.” We also saw Switch report record September sales at its Las Vegas SuperNAP facility, while Calxeda unveiled its product roadmap alongside its partnership with Penguin Computing.
On the weekend, Etisalat partnered up with Pacific Controls in order to provide cloud services for its enterprise clients, while IBM announced that IT outsourcing and data center services provider Academica selected IBM PureSystems to power its newest data center. We were also granted some insight into just how difficult it is for the health care industry to deal with data centers, and the sharing of data.
Perhaps the biggest story of the weekend came to us on Sunday, when HP announced to the Middle East what it termed “the industry’s first open-standards-based software-defined span infrastructure.” This new suite of technologies “covers all layers by including an SDN controller and SDN applications, services, and solutions that further strengthen HP’s Virtual Application Networks strategy.”
“In the cloud era,” said HP’s Middle Eastern Networking Director Fawwaz Qadan, “clients need a single point of control for the entire network, which enables them to deploy any application or service directly to the user within minutes. Only HP provides clients with a complete software-defined solution that automates manual configuarion tasks across hardware, software, and applications and from data center to desktop through a single control plane.”
We also saw Colo5 announce a partnership with Appcore and PeakColo select AMAX’s ServMax line of servers for its IaaS platform, while OCZ Technology made plans to display their Enterprise storage solutions at the Gartner Symposium/ITExpo, which ran from October 21st to October 25th.
Moving to Monday, the biggest story of the day (and, indeed, the week) was a crippling Amazon Web Services outage, which was traced back to a glitch at Amazon’s Ashburn, Virginia data center. A “small number of volumes” in Amazon’s Elastic Block Service began experiencing downgraded performance around 10:30 AM, and by 11:11 AM, it was obvious that there was a serious problem. Fifteen minutes later, websites across the board began to experience an outage, with sites such as Reddit, Pinterest, Minecraft, and Heroku all experiencing severe service problems. Amazon’s chief rival, Rackspace, is sure to capitalize on this.
Ultimately, it was just an unpleasant reminder that no hosting service is perfectly stable and reliable. Every cloud can crash.
Monday also brought us word of the development of a new data hub in Manhattan, at 325 Hudson Street. The carrier-neutral facility is being developed by Jamestown Properties, Amerimar Enterprises, and Hunter Newby, and will be used as an anchor for a new phase of data center development at the property. We also caught wind that HP invested $300 million into the research and development of the HP ProLiant Gen8 before releasing the server.
A new Gartner study released detailing the five major issues plaguing data centers was featured in a CyrusOne newsletter. These issues include an 800% growth in data over the next five years, space, power, and cooling consumption, increased pressure on performance due to virtualization, the modernization of legacy applications, and the identification and translation of business requirements.
Other news from Monday: Ignition Park readied itself for its first clients, which should arrive next month; Google’s data center spending rocketed up to $872 million in 2012’s third quarter, Core Services Corporation opened its fourth data center, located in Phoenix, Arizona; NYSE Euronext re-examined its UK data center in hopes of increasing the value of one of its most underutilized assets, the Moscow Exchange launched a connectivity point in Equinix’s London LD4 Data Center, Mellanox announced the creation of Mellanox Federal Systems, a subsidiary responsible for business development in federal government agencies, and No Limits announced the release of RaMP Version 2.3.
Moving into Tuesday, a report by Cisco projected that cloud traffic will grow six-fold by the time 2016 rolls around, exploding from 2011’s figure of 683 exabytes to reach a total of 6.6 zettabytes by 2016. To put that massive, almost unfathomable volume of data into perspective, Cisco explained that it is equivalent to roughly 92 trillion hours of streaming music. That’s somewhere around ten billion years of music. This was a part of Cisco’s latest Global Cloud Index, a release which has puzzled many, as it stands to aid Cisco’s competitors. Peerless Network opened up a new data center in Downtown Chicago, while Pacnet announced a series of new appointments to its executive team, part of a new strategic plan by CEO Carl Grivner to help the organization develop a refined business model and fast-track its growth.
The other big story of the day involved Intel and OnX, who announced their participation in the test-run of a new trading technology: a social network geared towards the Finteligent Trading community.
Other news from Tuesday: HP announced the installation of its Converged Storage and Critical Advantage Services at Croon Elektrotechniek B.V’s data center, Zamtel detailed plans to turn the Mwebmbeshi Earth Station into a data recovery center, Niobrara Data Energy Park released a set of promising metrics which validate the environmental soundness and sustainability of its Greenfield Energy Project, ContinuityX deployed its cloud-based solution portfolio in Cologix’s Dallas and Toronto data centers, and over in Vienna, a group of scientists shared a number of staggering advances in fields ranging from big data and cloud computing to biotechnology.
Wednesday saw T-Systems making plans to construct a massive cloud data center in Germany. The facility, which boasts an area of over thirty soccer fields, is designed to help Deutsche Telekom Subsidiary T-Systems meet the massive demand the cloud is seeing in Germany, and shall be constructed in Biere, Saxony-Anhal. Speaking of data centers, Wired gave us a peek inside Google’s Hamina, Finland facility, which includes, of all things, a sauna.
Wednesday also brought us a rather interesting message from the Green Grid: namely, that data center operators need to run their servers hotter, and damper, stating that the preconceived notion about servers only operating within a very narrow environmental tolerance is false; while Dell, after developing the “Zinc” server – an ARM-based box based around Calxeda’s EnergyCore processors – donated one to the Apache software foundation. This, says Dell, will allow the ASF to tweak and tune the Apache server along with several key components of Hadoop.
Other news from Wednesday: ICF International won an $18 million contract with the Department of Homeland Security, which will see the organization working with both local and public safety agencies to better protect the department’s Fusion centers; Schneider Electric announced the launch of a new suite of DCIM software, the EPA announced a number of changes to their Energy Star Program, the Niobrara Data Center and Energy Park opened its doors to developers, and an article on MarketWatch detailed the threats Cisco currently faces to its core router business.
Thursday dawned on InterNAP’s plans to build a new facility in the New York Metro market. The 100,000 square foot facility will be located in Secaucus, New Jersey, and marks NAP’s third data center in the New York market, adding 45% more capacity for NAP to address rapidly growing demand in the area. We also heard that Cognizant wishes to invest $25 million in data centers in the US and UK, in order to open new facilities in Arizona, Virginia, Amsterdam, and Slough, while Facebook adapted their Open Compute Project for colocation space.
A look at Lulea in Northern Sweden (location of one of Facebook’s recent data centers) gave us an idea of the long-term impact of a data center on a region’s economy, Online Tech added several new executives to its team, and web host iWeb started work on a Montreal data center expansion project.
Other news from Thursday: DuPont Fabros Technology’s shares slid downhill after Wall Street Analysts expressed displeasure regarding lease extensions with several of the organization’s largest customers, Atlantic.net achieved SSAE16 Certification, we got a look inside one of Facebook’s energy efficient data centers, a new telecommunications network entered the planning stages, with the intent to link Laos and Thailand; and finally, more rumors swirled about regarding Microsoft’s potential use of fuel cells at its new Wyoming data center.
Recent Comments