Monday, August 19th, 2013

Time	Event
10:52a	Enterprise Chef Reflects Expanding Vision for Opscode Infrastructure management specialist Opscode has realigned its offerings to reflect the company’s growing focus on enterprise customers. The Seattle company today announced Enterprise Chef, a single brand that includes both the hosted and on-premises version of Chef software. The rebranding marks a continuing expansion of focus for Opscode, which gained early traction in hyperscale computing, but now sees a big opportunity to help enterprises overhaul their IT operations. The reason is simple: software-driven automation offers clear benefits that are being adopted by a broader cross-section of companies. “Enterprise organizations are in the midst of a major business transformation, driven by the radically new way in which customers are purchasing and consuming goods and services today,” said Adam Jacob, Co-Founder and Chief Customer Officer of Opscode. “As a result, technology is serving as the key touch point to users and the role of IT has shifted from the back office to the front office. Today we’re delivering an automation platform that accelerates this transformation by delivering on-demand IT services to achieve the speed necessary for meeting the new expectations of customers.” Automating All the Things Chef is an open source framework using repeatable code – organized as “recipes” and “cookbooks” – to automate the configuration and management process for virtual servers. It enables users to deploy infrastructure as code across any operating system from Windows to Unix and Linux, across physical, virtual or cloud infrastructures. Enterprise Chef combines two separate products – Private Chef and Hosted Chef – which have now been realigned under a single brand, available as on-premise software or as a hosted service. The shift isn’t only aspirational. Opscode says that Fortune 1000 companies now represent 60 percent of its sales, with the Fortune 1000 customer base having grown 150 percent in the past year. New Features for Networking, Storage To continue that momentum, Opscode also announced new collaborations to extend Chef’s capabilities in networking and storage. That includes partnerships with Arista Networks, Cisco Systems, Cumulus Networks, Juniper Networks and Plexxi to integrate Enterprise Chef into next-generation networking technologies. “The enterprise shift to scale-out computing has created a massive management challenge for the networking layer, with IT operations teams often employing only one networking person for thousands of ports,” said Doug Gourlay, vice president Systems Engineering and Technology Marketing at Arista Networks. “Eliminating manual port configuration is the first, and most critical, step in solving the networking dilemma in today’s data centers. Combining Chef and Arista EOS provides a flexible solution for abstracting management into simple code commands that ensure bandwidth availability and consistency at any scale.” Opscode also announced a storage initiatives, teaming with Netherlands-based IT outsourcing leader Schuberg Philis and software-defined storage leader Nexenta to deliver a Chef Cookbook for automating NexentaStor ZFS-based storage resources. Windows Friendly For enterprises using Windows, Opscode continues to work with Microsoft. Today the companies said Chef can now take advantage of the new Windows PowerShell Desired State Configuration feature of the Windows Management Framework (WMF). “As enterprises look to IT as a catalyst in the race to market, these organizations must automate tasks in compute infrastructure that includes Windows-based environments,” said Christopher Brown, CTO, Opscode. “Opscode Chef is the only open source automation platform with native Windows functionality, which can help customers drive faster, less risky product and application development cycles. Our continuing collaboration with Microsoft ensures Windows Server administrators can leverage the entire scope of Chef’s open source automation capabilities for more rapidly delivering goods and services.” And that, Jacob says, is the goal for enterprises ready to embrace agile software development, and seeking the tools and talent to do so. “Software and culture go together,” says Jacob. “The external market -the world at large – is pushing this trend. People want to ship software whenever they need to without drama. It’s a question of how long it takes, and one of the key drivers is whether you have the right software to build your company.” “Chef works best when you’re automating big business problems,” Jacob added. “That’s where Chef really shines.” (Comment on this)
12:21p	Twitter’s New Infrastructure Weathers Massive TweetStorm Monitoring data from Twitter shows a huge spike in Tweets per second during a 10-second period on August 3. In an unusual test of Twitter’s infrastructure, Japanese anime fans set an all-time record for Tweets per second earlier this month during the showing of a 27-year-old movie on television. The entire event lasted less than ten seconds, but showcased Twitter’s new and improved back-end, which has improved performance while vastly reducing the number of servers needed to run the microblogging service. On August 3, Twitter users in Japan watched an airing of “Castle in the Sky,” a 1986 movie from master anime filmmaker Hayao Miyazaki. Fans of the movie observe a tradition of posting a key line of dialogue in online forums at the exact moment it is spoken during the film. The practice has shifted to Twitter, and on August 3 resulted in an all-time one-second peak of 143,199 Tweets per second, shattering the previous record of 33,388 Tweets per second on New Year’s Day 2013. “To give you some context of how that compares to typical numbers, we normally take in more than 500 million Tweets a day which means about 5,700 Tweets a second, on average,” wrote Raffi Krikorian, Twitter’s VP of Platform Engineering, in a blog post about the event. “This particular spike was around 25 times greater than our steady state.During this spike, our users didn’t experience a blip on Twitter.” New Architecture Vanquishes Downtime So where was the Fail Whale? Twitter weathered the Tweetstorm due to a retooling of its software and infrastructure after a series of high-profile outages during the 2010 World Cup. The service’s Rails-based infrastructure was constrained in storage and throughput, and unable to scale to the massive traffic from soccer enthusiasts. “We were ‘throwing machines at the problem’ instead of engineering thorough solutions,” Krikorian wrote. “Our front-end Ruby machines were not handling the number of transactions per second that we thought was reasonable, given their horsepower. From previous experiences, we knew that those machines could do a lot more.” Twitter’s Rails servers were “effectively single-threaded” and only capable of 200 to 300 requests per second per host, Krikorian explained in a detailed technical overview of the project. “Twitter’s usage is always growing rapidly, and doing the math there, it would take a lot of machines to keep up with the growth curve.” Twitter instead began switching workloads to a revamped system using JVM (Java Virtual Machine), and had big goals. “We wanted big infrastructure wins in performance, efficiency, and reliability – we wanted to improve the median latency that users experience on Twitter as well as bring in the outliers to give a uniform experience to Twitter,” writes Kirkorian. “We wanted to reduce the number of machines needed to run Twitter by 10x. We also wanted to isolate failures across our infrastructure to prevent large outages – this is especially important as the number of machines we use go up, because it means that the chance of any single machine failing is higher. Failures are also inevitable, so we wanted to have them happen in a much more controllable manner.” Fewer Servers, Better Reliability A significant benefit of the overhaul, according to Krikorian, is the reduction in servers required to run Twitter. “Twitter is more performant, efficient and reliable than ever before,” he writes. “We’ve sped up the site incredibly across the 50th (p50) through 99th (p99) percentile distributions and the number of machines involved in serving the site itself has been decreased anywhere from 5x-12x. Over the last six months, Twitter has flirted with four 9s of availability.” Twitter still needs plenty of server space. It recently began a major expansion of its data center infrastructure that includes large footprints at RagingWire iNS acramento and QTS in Atlanta. For those seeking even more details about Twitter’s back-end overhaul, check out this 2012 video in which Krikorian provides a deeper dive on the project: (Comment on this)
12:30p	Four Myths About Automated Disaster Recovery Ralph Wynn, technical director of product marketing for FalconStor Software, is a storage professional with more than 15 years of experience in product management, marketing, support and deployment. Prior to joining FalconStor, Ralph worked at Bocada, Synscort and Symantec.. RALPH WYNN FalconStor In the months since Hurricane Sandy struck the Eastern United States, disaster recovery (DR) planning has been covered in the media and discussed by IT administrators from almost every angle. Every company knows it needs to prepare for literal storms and figurative ones – the instances of human error, malicious acts and equipment failure that can bring IT down just as surely as a superstorm. DR, data protection and accessibility are 24-hours-a-day, seven-days-a-week, year-round critical tasks for IT. In fact, data center managers now understand that they must look beyond just protecting and backing up data; they need to protect complete IT services on which the business depends. The challenge comes in how to implement DR successfully into today’s increasingly complex environments, since traditional data protection methods cannot scale easily to handle massive data growth. A DR Insurance Policy for Every Data Center The complexity of today’s data center creates a time crunch that is only getting more severe. DR is the most time-consuming process within the data center, and it is often pushed to the bottom of daily IT priority lists. As data center managers deal with mounting numbers of end-user requests and other daily projects, they understandably lose focus on the just-in-case processes that aren’t as pressing. And yet, just as we all maintain insurance policies on our homes to protect us from the unthinkable, data centers must have DR plans to do the same job. DR is the IT insurance plan, and it protects the data, the applications, the systems, and the infrastructure that make up critical IT services that runs the business. Simply put, DR is critical for continuity of business operations. Companies can lose millions of dollars when their data centers go down. This cost comes in the form loss of productivity, damaged corporate reputations and other repercussions. Most companies state they can’t afford downtime that lasts for more than four hours without incurring painful losses. IT administrators are all too aware of that pressure, so when a disruption occurs, they scramble to execute the complex, stressful series of steps required to get up and running. So, how can IT leaders alleviate that stress and get back to business more quickly after a disaster? How can data center managers ensure that their IT insurance plans are adequate? The answer lies in proper planning, testing and implementation of DR plans and solutions that are designed specifically for today’s complex, data-intense environments. In the past, too many DR plans failed because they weren’t executed correctly, conventional data protection technology couldn’t scale to protect IT services, and the plans weren’t thoroughly tested. Automated DR addresses all of these pitfalls by mechanizing the traditionally time-consuming manual recovery process. With automated DR, companies can recover systems within minutes, rather than hours or days, with a minimum of hands-on work. Dispelling the Myths about DR Automation Automated DR delivers time savings and decreases costly downtime, so why aren’t more data centers embracing it? The answer is simple: misunderstanding. The best way to combat the lack of understanding is with information, so let’s look at the four most common myths that inspire distrust in automated DR: 1. Myth: Automated DR cannot be customized to the environment. IT administrators believe that automated DR solutions are not able to handle the different kinds of hardware or physical and virtual servers used by most data centers today. In many cases, IT admins have to scramble to recover systems quickly and need to make use of any systems, physical or virtual, that are readily available. Reality: There are automated DR solutions that can failover and failback between dissimilar hardware as well as from physical-to-virtual, virtual-to-virtual, or virtual-to-physical systems, allowing for quick recovery in any environment. Regardless of the type of servers, hardware or connectivity, automated DR can handle replication and recovery within heterogeneous environments. 2. Myth: Automated DR doesn’t allow for full testing. There is a perception that automated DR lacks the mechanism to test an end-to-end recovery scenario. There are some solutions that only offer the ability to test the recovery of a particular data set, application or part of the process. Without a full test, IT managers don’t feel confident about the ultimate disaster system in the event of an actual emergency. Reality: A full, in-depth test of the entire environment is critical and must be done – at a minimum once per quarter – to ensure all the processes work. When selecting an automated DR solution, IT managers must examine the testing procedure in detail to ensure they can run a full test without causing significant downtime. 3. Myth: Automated DR requires multiple sources for data recovery. There are some DR automation solutions that can only restore part of the data on the system and need to rely upon other, potentially slower, backup applications for a full restoration. For example, there is a concern that if it has been four hours since the point of failure, the automated DR solution may only be able to recover two hours of lost data. The remaining data would need to be pulled from a secondary backup source. These types of solutions greatly hinder recovery time objectives. Reality: A true automated DR system goes back to the original point of failure. Automated DR technologies that are integrated with disk-based continuous data protection and snapshot technology will allow companies to fully recover in one complete process. 4. Myth: Automated DR can only recover one system at a time. IT admins believe that automated DR solutions can only recover one system or application at a time, which is not adequate for an outage that spans multiple systems, applications and services. This concern stems from some products that purport to be automated DR solutions but can only recover one server at a time as part of the recovery process. Companies that experience failures that affect the entire data center are left running numerous time-consuming single-system restores. Reality: Today’s advanced automated DR solutions allow for multiple machines – up to five systems at once – and complete IT services to be restored to full operation in a matter of minutes. This service-oriented approach to recovery focuses on the integrated constellation of systems, applications and data that make up critical data center services to eliminate or significantly reduce downtime. Automated DR is a company’s insurance plan to eliminate costly downtime and stem productivity losses when systems crash. IT admins can remove the limitations and failures of traditional data protection solutions in the data center, but many hold to these common myths about automated DR. IT managers can overcome their distrust of automation by evaluating all available solutions on the market. They will find that the best of today’s automated DR systems are much more capable and reliable than they currently believe. 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)
1:56p	Digital Realty Adds Cloud Connectivity Services in UK Digital Realty is offering enhanced cloud connectivity services in the UK, in partnership with communications service provider Epsilon. By leveraging Epsilon’s outsourced connectivity model and expertise, Digital Realty (DLR) will offer a dynamic bandwidth solution to cloud and content providers for seamless backbone connectivity throughout its London-area data centers. It begins with an initial two Points of Presence (PoPs) in two Digital Realty data centers, connecting data centers in Chessington and Woking with 22 key London-area Internet and metro gateway centers. The company will extend this across Europe. “Our goal is to provide flexible solutions that accommodate the evolving needs of our customers’ businesses, now and in the future,” said Bernard Geoghegan, Managing Director, EMEA, for Digital Realty. “Our London-area customers require on-demand and rapidly scalable automation of their bandwidth as part of their data center solutions. We are pleased to offer this platform as we continue to build the Digital Realty Ecosystem across our portfolio in Europe and globally.” Digital Realty is leveraging Epsilon’s experience in operating high-density metro connectivity platforms. The new platform brings enhanced connectivity for cloud and content providers in Digital Realty’s London-area data centers, with seamless interconnectivity to over 65 data centers globally. This enhanced connectivity platform will operate on a major dark fiber ring commissioned by Digital Realty. Customers will benefit from connectivity to 25 additional London metro gateways, as well as interconnectivity to more than 500 carrier, service, internet exchange, cloud, content and local access providers currently connected on Epsilon’s exchange. “Epsilon’s global network exchange, powered by BTI’s Cloud Networking Solution, provides the platform that we believe will help us achieve our distributed cloud data centre vision,” said Robert Bath, Vice President of Engineering, at Digital Realty. “We expect the solution to attract a new class of global cloud and content provider customers to our data centers.” Epsilon has created an application-aware network to rapidly scale capacity to support 100G connectivity, and facilitate on-demand switching of cloud based application workloads between data centers. “We see growing demand from data centre customers for our outsourced connectivity model and this agreement with Digital Realty reaffirms our commitment to supporting cloud and content connectivity development,” said Andreas Hipp, Chief Executive Officer at Epsilon. “As a global provider, we are driven by service excellence and agile operational objectives. Working with BTI helps us achieve our dual mission by simplifying network operations and increasing service velocity, while enabling the introduction of new cloud connectivity services.” (Comment on this)
4:07p	QTS Confirms Plans for IPO, Major Expansion A view of the rows of racks packing a pod inside the QTS Downtown Atlanta data center. QTS (Quality Technology Services) has confirmed plans for a $300 million initial public offering (IPO) that will help fund an expansion of the company’s footprint, focused on its massive data centers in Dallas, Richmond and Atlanta. In an SEC filing, QTS outlined its plans to go public on the New York Stock Exchange, trading under the symbol QTS. The company plans to convert to a real estate investment trust (REIT) and operate as QTS Realty Trust Inc. QTS has big ambitions for the future, fueled by its strategy of buying massive industrial facilities and adapting them for data center use. In its SEC filing, the company said it plans to expand seven of its data centers across the county, investing up to $277 million to add more than 312,000 square feet of customer space in key markets over the next two years. The majority of that growth is projected for Dallas ($88 million investment), Richmond ($78 million) and Atlanta Metro ($54 million) , but QTS also plans to add more space at existing sites in Jersey City, Sacramento, Santa Clara and Suwanee, Georgia. Focused on Redevelopment “We believe our redevelopment pipeline provides us with a multi-year growth opportunity at very attractive risk-adjusted returns without the need to construct new buildings or acquire additional properties or land for development,” the company said. “We currently target a stabilized return on invested capital of at least 15% on average for our redevelopment projects.” QTS operates 10 data centers in seven states offering 714,000 square feet of raised floor data center space and 390 megawatts of available utility power. The portfolio also includes 2.2 million square feet available for redevelopment, including 1.1 million square feet that can be converted to raised floor data center space. On the financial front, QTS reported revenue of $84.4 million in the first half of 2013, with net income of $7.1 million and funds from operation (FFO, a key benchmark for REITs) of $26.7 million. In 2012, the company had revenues of $157.6 million and FFO of $45.2 million. Here’s a look at some of the additional opertating details highlighted in the company’s IPO filing: The greater Atlanta market, where QTS is the dominant player, represents a substantial piece of the company’s  business  Of the $52.9 million in revenue from data center leasing in 2013, $38 million is generated from the company’s two Atlanta facilities, Atlanta Metro ($25 million) and Suwannee ($13.4 million). The campuses in Santa Clara ($5.5 million) and Richmond ($4.7 million) are next, with the other four sites totaling $4.3 million in revenue. QTS has some lease renewals ahead. Leases representing approximately 27 percent of the company’s raised floor and 44 percent of its annualized rent will expire in either 2013 or 2014. That includes a number of month-to-month leases. As of June 30, QTS had 870 customers, with none accounting for more than 8 percent of monthly recurring revenue (MRR). About 39 percent of recurring revenue is from customers using managed hosting or cloud offerings. Fortune 1000 and equivalently sized private and/or foreign companies accounted for approximately 57 percent of MRR. QTS Realty Trust will be created through a “formation transaction” that will consolidate holdings between current investors, including the private equity firm General Atlantic and Chairman and CEO Chad Williams. The business unit that provides managed services to customers, Quality Technology Services Holding LLC, will operate as a subsidiary. The company will go public using a two tier stock ownership that preserves voting control over the company in a class B preferred stock. No date was provided for when QTS plans to commence trading. (Comment on this)
5:00p	Tegile Raises $35 Million For Rapid Expansion Hybrid storage array provider Tegile Systems announced it has closed on a $35 million Round C funding led by venture firm Meritech Capital Partners with additional investment by original stakeholder August Capital and strategic partners Western Digital and SanDisk. Presented as a ‘scaling round’, the funds will enable Tegile to rapidly accelerate the expansion of its proven technology and successful go-to-market strategy by utilizing the capital raised to further strengthen its sales and customer support efforts throughout its North American and European channels and to accelerate its new product/technology development. Tegile has been aggressively building the company from the outset to become the only one having generated more revenue than it has taken in outside financing.  Before opening its Round C funding, the company improved its sales and revenues from doubling every two quarters to doubling every quarter. “The time has come for us to scale the financial success we’ve thus far been afforded and leverage these opportunities for even greater revenue prospects and increased sales,” said Rohit Kshetrapal, CEO of Tegile. “The financing we’ve secured in this latest funding round allows us to do just that.  We are pleased that our Round C investors include two strategic partners that will strengthen both our relationship and product roadmap even further, and working with an investment firm known to quickly move technology companies to the forefront. We look forward to Paul joining our board.” Investor SanDisk noted that the Tegile solution provides better price-performance than traditional all-hard-disk-drive storage arrays and offers a compelling way to address evolving enterprise requirements in a cost-effective manner. Through its strategic partnership with SanDisk, Tegile will be able to leverage SanDisk’s expertise and capabilities as one of the world’s leading innovators of Flash memory technology. “With our unique flash-based hardware and software solutions and vertically integrated business model, SanDisk is helping enable a new generation of companies transform the enterprise,” said Sumit Sadana, executive vice president and chief strategy officer at SanDisk. “We believe that Tegile is well positioned to bring the significant benefits of flash technology to enterprise customers through their innovative hybrid system solution.” (Comment on this)
9:35p	Amazon Recovers from Brief Outages for Retail Site, Some AWS Services Amazon says it has resolved issues that briefly caused downtime for its flagship Amazon.com retail sites, along with about 45 minutes of lost functionality for some of its Amazon Web Services (AWS) businesses. Amazon said it experienced increased error rates for APIs for its Elastic Compute Cloud (EC2) service in its US-EAST-1 region between 11:45 AM and 12:30 p.m. Pacific time, the company reported on its status dashboard. “The issue has been resolved and the service is operating normally,” the company reported. Significantly, the problems prevented some customers from accessing the AWS Managament Console. Amazon’s Cloud Formation, Elastic Beanstalk, Relational Database Services and Mechanical Turk services also reported temporary access issues. really glad we all survived that AWS API outage. proud of all of you. how about a group hug? — Jay Gordon (@JayGordonOps) August 19, 2013   While AWS has periodic outages, downtime is rare for the Amazon home page and retail site. Amazon.com was returning a “Website Temporarily Unavailable” message briefly, but appeared to have recovered by about 3:30 pm Eastern. (Comment on this)

<< Previous Day

2013/08/19 [Calendar]

Next Day >>