Thursday, February 19th, 2015

Time	Event
11:00a	Couchbase Integrates NoSQL Database With Hortonworks Hadoop NoSQL database provider Couchbase announced new product integrations with the Apache Hadoop ecosystem. Couchbase has integrated Hortonworks Data Platform (HDP) 2.2 into its Couchbase Server. Apache Kafka brings in distributed messaging, and Apache Storm integration brings in stream processing in the setup. The goal of the integration, driven by Couchbase, is to provide operational and analytical requirements with a single solution.The Kafka connector strengthens the streaming data between Couchbase and Hadoop. Couchbase is the operational store and Hadoop is the deep analytics. The big data market can be segmented into two halves: operational data management, where Couchbase and NoSQL vendors play, and analytical data processing, which has been led by Hadoop vendors. Anil Madan, senior director of engineering at Paypal, a Couchbase customer, spoke about the power of the integration in a release. “As the operational data store, Couchbase is easily capable of processing tens of millions of updates a day, or more,” he said. “Streaming through Kafka into Hadoop, these key events are turned into business insight and that’s why we are very excited about this partnership.” Kafka uses Couchbase Server’s Database Change Protocol (DCP), new to 3.0, to stream data from Couchbase Server to the message queue in real time. Apache Storm then consumes the messages for real-time analysis. The data is then written to Hadoop for further processing, with analysis written to Couchbase Server for access by real-time reporting and visualization dashboards. “As more data streams flow into enterprises from a larger variety of sources, IT leaders must extend data processing to include more use of push (event-driven) models, “said Nick Heudecker, Gartner, in a release. “The opportunities and the risks of not harnessing high-velocity data streams are too significant to ignore. This shift in data models exposes new business and operational opportunities in information security, the Internet of Things customer engagement, fraud detection and telematics, among others.” For offline analytics, Hortonworks certified the Couchbase Server plugin for Sqoop to support bi-directional data transfer bettwen Couchbase Server 3.0.2 and HDP 2.2. This lets enterprises export operational big data stored in Couchbase to HDP for offline analysis and refinement, then import back into Couchbase Server. “Today’s enterprises need to build web, mobile and IoT applications that deliver contextual insight in real time,” said Rod Hamlin, vice president of business development at Couchbase, in a release. “The work we have done makes it faster and easier for IT organizations to deploy real-time big data architectures that enable them to analyze data continuously and support real-time decision making and personalization.” Couchbase is the result of a 2010 merger between CouchOne and Membase. In 2012, it shifted focus away from Apache CouchDB into its own NoSQL database technology, which is also open source. Another CouchDB player is Cloudant, which was acquired by IBM. Cloudant had its own version of CouchDB called BigCouch and still contributes to Apache CouchDB. Couchbase raised $60 million last June, noting sales were up 400 percent. In August, it partnered with CumuLogic, ElastiBox and Cloudsoft to help businesses deploy Couchbase Server to a cloud of choice. (Comment on this)
12:45p	HP to Ship Commodity Switches With Cumulus Linux OS HP has become the latest “legacy” IT vendor to announce it would ship commodity switches for web-scale data centers that support network management software other than its own. The Palo Alto, California-based company said its differentiation in the competitive commodity switch market would be a global supply chain that extends across more than 160 countries and a smorgasbord of support and consulting services. Internet giants, such as Google, Facebook, and Amazon, which operate massive data centers around the world, have found it more effective to design their own hardware and have so-called “Original Design Manufacturers,” the likes of Taiwan’s Quanta and Foxconn, manufacture it for them. The trend has created a problem for incumbent IT vendors, such as HP, Dell, IBM, and Cisco, which found themselves competing with contract manufacturers of their products for the same high-volume deals. Facebook, through its Open Compute Project, has spread awareness about cost effectiveness of this IT procurement model, and there is now growing interest in low-cost commodity hardware among enterprises who are not necessarily Internet giants, creating a new threat to the incumbents’ market share. Facebook started with servers, but now, it has also designed its own network switch and plans to contribute that design to OCP like it has with its server designs. HP’s competitors Dell and Juniper have already announced open commodity network switches of their own. Dell said it would ship data center switches with a Linux-based network operating systems by Cumulus Networks or by Big Switch Networks, as alternatives to its own network OS, last year. Juniper announced it would start shipping a commodity switch that would support any OS sometime in the first quarter of 2015. Taiwanese manufacturer Alpha Networks will manufacture its “white box” switch. Notably, Cisco has not introduced commodity switches. The world’s largest data center networking vendor has built an empire selling tightly coupled hardware-and-software bundles, and cheap open network hardware is a threat to its dominance. HP’s commodity switches will come with Cumulus software as well. They will be manufactured by Taiwanese ODM Accton Technology Group. The company plans to start shipping switches for 10G/40Gspine and 10G leaf data center deployments in the second half of the year. It expects to expand the line with 25G, 50G, and 100G switches, as well as 1G, sometime in the future. “Web-scale data centers are a growing market segment with distinct needs for high scalability and network flexibility,” George Tchaparian, general manager of data center networks at Accton, said in a statement. “This joint venture will build on the expertise of both organizations to provide global customers with a unique disaggregated networking solution built on new HP-branded brite box switches and choice of networking operating system software.” (Comment on this)
1:00p	Why Cloudera and MapR Opted Out of Pivotal’s Hadoop Consortium When it announced it would open source its entire suite of big data analytics tools earlier this week, Pivotal also announced formation of an industry group that would bring what the EMC- and VMware-owned software company said was a much needed common set of standards to ensure compatibility across technologies in the Apache Hadoop ecosystem. Founding members of the group, called Open Data Platform, include enterprise Hadoop heavyweight Hortonworks (which now has an expansive partnership with Pivotal), GE, which owns a 10-percent stake in Pivotal, as well as EMC and VMware. The list also includes numerous non-affiliated heavyweights, such as IBM, Verizon, CenturyLink, Capgemini, Teradata, and a couple of others. Glaringly, two of Hortonworks’ biggest competitors in the enterprise Hadoop market – Intel-backed Cloudera and MapR, which last year closed a $110-million funding round led by Google Capital – were not part of the ODP announcement at all. Sundeep Madra, who leads Pivotal’s Data Products Group, told DCK he’d love for them to join. But top execs at both companies told us they don’t want anything to do with ODP. They said the initiative seemed self-serving, and its stated goal of creating a “common core” for Hadoop was redundant, since a framework for interoperability already existed within the Apache community itself. “We haven’t really seen interoperability across projects as a major issue at all,” Jack Norris, chief marketing officer at MapR, said. “We think the market, and especially the Apache Software Foundation, is working fine. That’s why it seems like [ODP] is more about a partner program than a community initiative.” MapR management were approached about joining the ODP but declined. “We had the chance to participate and basically decided not to.” Mike Olson, a Cloudera co-founder and the company’s chief strategy officer, said ODP was a “marketing ploy,” and a “mistake.” He agreed that Apache ensured interoperability in the Hadoop ecosystem aptly. “The APIs and services of HDFS [Hadoop Distributed File System] and MapReduce have been carved in stone for years,” he wrote in an email. “There’s no incompatibility among those APIs from any vendor. Zero.” Cloudera management, Olson said, learned about the pending ODP announcement about one week ago and approached Pivotal to learn more. Once they did, they decided not to join. “Pivotal and Hortonworks claim that the ODP is driven by an industry-wide longing for standardization in the Apache Hadoop ecosystem,” he wrote in a blog post explaining his company’s position. “I don’t believe them.” Cloudera has about 1,450 companies in its partner ecosystem, and its management has not heard from that ecosystem that there were companies that were confused about building applications on core Hadoop, Olson wrote. His biggest objection to ODP was that it seemed like a club for companies with deep pockets that’s very different from the Apache community of developers that drive Hadoop forward. “Developers in Hadoop already collaborate to design and implement standards, using tried, tested, successful open source collaboration tools,” he wrote. Notably, one of the announcements Pivotal made this week was that its analytics tools would soon support Hortonworks’ Hadoop distribution. Pivotal has its own distribution and said it would continue to support it along with Hortonworks’. (Comment on this)
4:30p	The Data Flow Imperative: Fighting Barriers to Promote Growth Scott Belcher is the CEO of the Telecommunications Industry Association, the leading association representing the manufacturers and suppliers of high-tech communications networks. As nearly every aspect of government, commerce, and individual life becomes increasingly digitized, the benefits of an all-IP world have become essential to consumers and the American economy. But a growing public policy movement by countries around the globe holds the potential to dramatically impact the future of communications here at home. Barriers to Flow of Information Individual countries are considering – and in some cases, implementing – policies that will create significant barriers to the cross-border flow of information. In addition, there is a trend to “localization,” in which countries require that data and data centers be located within their borders. These types of regulations threaten significant economic damage by limiting companies’ ability to sell products and conduct business overseas, and they hold the potential to create a balkanized Internet, reducing the many benefits of its open architecture. To be sure, the U.S. will feel a direct, and significant impact if this movement becomes widespread. According to the U.S. International Trade Commission, “digitally intensive” U.S. businesses sold over $935 billion in products and services and purchased $471 billion in 2012. Data flow regulations will create a nearly impossible landscape for these innovative, data-driven companies. The ability to send commercial data across borders with minimal restrictions is vital for U.S. companies of all sizes, whether they are conducting business here at home or around the world. And it’s not just the tech industry that will be impacted – restricting data flows will put at risk a wide range of business sectors. Small retailers, for example, are increasingly reliant on e-commerce as an essential way to expand their businesses and access global markets. Tackling the Issue Head-On With so much at stake, ensuring that data flows can move across borders without undue interference or restriction must be top priority for policymakers and the Obama Administration. It starts with sending a clear signal to our trading partners that our nation is prioritizing this issue, and is committed to fighting the global trend toward commercial data restrictions and data center localization. In addition, we must promote the use of voluntary agreements that allow for the implementation of collaborative, multi-stakeholder solutions to data security and privacy. We’ve already seen how effective such agreements can be in protecting consumers, while avoiding unnecessary and heavy-handed government regulatory schemes. For instance, both the U.S.-EU Safe Harbor, administered by the U.S. Department of Commerce, and the APEC Cross Border Privacy Rules System have built trust in the cross-border transmission of personal information by providing a mechanism for businesses to demonstrate that their privacy policies meet European Union or APEC Privacy Framework requirements. These agreements have proven effective for data flows between the U.S. and Europe, which are the highest in the world according to the Brookings Institution. However, the growing economic might of Asia and Latin America, combined with the global nature of the Internet, provide a strong impetus to further develop the Cross Border Privacy Rules System so that it eventually includes all 21 APEC member economies. The U.S. must also pursue strong open data-flow provisions in trade agreements that are now being negotiated – this includes avoiding both restrictions on where data can be stored and localization requirements for data centers or other related infrastructure. It is a good sign that these types of provisions are being negotiated in the Trans-Pacific Partnership (TPP) agreement, and we hope to see similar provisions in future trade agreements to which the U.S. is a party. An Engine of Consumer Empowerment By avoiding heavy-handed regulation and barriers to cross-border data flows, while promoting smart and collaborative data privacy agreements, we can help make sure that the Internet continues to be an engine of consumer empowerment and economic growth in the U.S. and around the globe. Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission processfor information on participating. View previously published Industry Perspectives in our Knowledge Library.   (Comment on this)
5:00p	What Should You Consider When Moving to Colocation or Managed Services? When your data center is reaching the end of its life, its power or cooling are strained, and it’s a struggle to keep pace with ever growing capacity demands, one starts to consider colocation or managed services solutions. Just buying some colo space is not as simple as it seems. How do you go about defining your needs, finding the right provider, negotiating the deal, and migrating to the space? Tad Davies, Senior Vice President, Bick Group, will speak to these elements involved in the whole process of moving to an outsourced provider, in the session titled, “Considerations on Colo (Managed Services)” at the spring Data Center World Global Conference in Las Vegas this April. To help attendees, Data Center Knowledge is doing a series of posts on the topics covered at Data Center World to bring more attention to the issues covered by session leaders, such as cost, capacity and cloud, among others. Bick Group, which has been in business for more than 50 years, provides consulting services to clients as well as providing facilities services. Davies oversees the client consulting side of the business. Why Go Colo? “Capacity is one of the drivers,” Davies said, about an enterprise’s move to a service provider. “Usually the main driver of colocation or managed services is the company’s asset is not functioning well. It’s at the end of life, such as power or cooling are insufficient, or it could be capacity.” This situation where one’s data center is no longer meeting the business need is a signal that colocation services might be in order. However, from Davies’ perspective, it’s not a one-size-fits-all approach. There are many factors to sift through and consider. “From my point of view, everything is situational. You have to do fact-based decision making,” he said. A firm’s data center strategy must address the particular needs of the business. Needs Assessment When looking at the service provider market, it’s important to first do your homework. “Hammering down what your needs are” is a vital step, according to Davies. He recommended that a client should think through questions such as, “Are you looking for colo?;”Are you looking for managed services, and if so, what kind?”; “What are the business needs?”; “Is the company in a growth or expansion mode?” and “Are there business growth issues on deck, such as potential acquisitions in the pipeline?” Different Financial Angle From the financial angle, the colocation and managed services approach is different from the “build your own” data center process. If you build a data center asset, the investment is over a longer time frame, such as 15 years or more, and the funding comes from the CapEx side of the balance sheet. “You look at your first need, five or six years ahead, and you get that funded because you don’t want to go back to the board for more funding,” said Davies. While for some organizations avoiding CapEx is desirable, for others there is not that incentive. The financial requirements depend on the individual enterprise, of course. But it’s good to recognize that the purchase of colocation space works differently, he said, in that the provider can deploy more capacity “relatively instantaneously.” This ability to buy space “as you go” can be very helpful. However, he noted, that while it is easy to “scale up,” it is not as easy to “scale down.” This is why an accurate needs assessment is critical. A Provider’s Business Model Matters Also, a key consideration in picking a provider is “understanding what their business model is,” said Davies. “You need to understand what is important to the provider in addition to what is important to you. If you know the service provider’s business model, you will know what levers you can pull,” he said. The Technical Match After you’ve defined your needs and assessed your approach on the financials, you look for providers that will meet your level of infrastructure and resiliency. “Most colocation and managed service providers have pretty good infrastructure today,” said Davis. “It’s a bit of an arms race in the service provider world.” But, he cautions, your organizaiton might not need the top of the line across all uses. The range of infrastructure choices available go from the most highly resilient, or fault tolerant, data centers to those that are simply “concurrently maintainable.” Another item to consider is understanding the IT organization’s applications and their dependencies. While most companies know what their inventory of assets and infrastructure and even applications, their weaker area is knowledge of application dependencies. It may take effort to determine what these are, a lot of planning in the migration phase, and this may prove to be one of the most challenging aspects of moving to colocation services. Migration One of the advantages of moving from an enterprise data center with a limited geographic footprint is that a colocation services provider often has the ability to access multiple, geographically diverse data centers. This can improve one’s back up and disaster recovery approaches by having primary and secondary sites operating. However, Davies pointed out, the degree of difficulty of migrating may reduce the number of options one has in colocation provider selection. Geographies can impact how one picks a provider, and how one ultimately, migrates to that provider. Migration can be highly complex, he said. Learn More To find out more about how to go about selecting the right colocation and/or managed services and moving to those sites, attend the session by Davies at spring Data Center World Global Conference in Las Vegas. Learn more and register at the Data Center World website. (Comment on this)
5:30p	Important Facts to Consider When Working With Remote Server Management As an IT administrator, lab, or network manager, you need to access and control multiple computer systems. You have routine maintenance, requirements for troubleshooting and even real-time monitoring needs. With that in mind, your remote server management tools must enable you to work quickly and securely to access any system regardless of type, brand or location. In this e-book from Raritan, we look at how server management tools fall into three general categories: Software-based remote access systems – RDP, VNC Computer embedded service processors – iLO, DRAC, RSA Out-of-band KVM-over-IP switches – Raritan Dominion KX2 While these categories describe how the tool itself is physically deployed, it’s critical to understand the real advantages and limitations of the tools in each category. As the paper outlines, there are four critical criteria points when it comes to your remote server management tools. Consider this: Reliability – Will my remote access and control work when I need it most? Security – Is my remote access and control tool safe from hackers and cyber-attack? Manageability – Is my remote access and control tool easy to manage and maintain? Power – Can I access and control all of my devices wherever they are and perform my day-to-day tasks? Remember, your data center platform will only continue to evolve. Your environment will take on new workloads, additional VMs, and new infrastructure components. Through it all, it’s critical to be able to control this environment and management whenever needed, whether remote or onsite. Download this e-book today to see how the three mentioned remote server management tools stack up and which is the right one for the job. You’ll understand the various use-cases, deployment methodologies and where each category ranks in terms of security, performance, reliability and manageability. (Comment on this)
6:00p	Nevada, North Dakota Contemplate Data Center Tax Breaks State-level legislation aimed at attracting data centers through tax incentives in Nevada and North Dakota is in the works. A new data center tax break bill has been introduced in Nevada (Senate Bill 170) and a North Dakota bill (House Bill 1089) has passed the State House of Representatives. Nevada is an established and growing market, while North Dakota is hoping to spur some initial activity. States continue to implement or contemplate tax breaks for data centers because of their ability to spark local technology industry. Data centers don’t create large amounts of jobs, but the jobs they do create are high-paying jobs. Local economies benefit from data center property taxes and from taxes on their massive ongoing power purchases. Colocation providers bring the benefit of attracting high-tech companies to the areas as their tenants, which also has positive economic impact locally. The Nevada bill would provide partial sales and personal property tax abatements for 20 years to data center operators investing $250 million and meeting employment requirements. These abatements would also extend to colocation facilities. There are several existing and growing data centers in Nevada, facilities by Switch in Las Vegas and by Apple in Reno being two of the more famous examples. Switch announced in January it would also build a data center in Reno, a $1 billion facility where eBay will be the anchor tenant. The North Dakota data center tax break bill would provide a sales tax exemption for equipment and software used in a qualified data center. North Dakota does not have a single facility that would qualify, so the bill is an attempt to attract data centers. Nevada is examining the potential benefits of tax abatements for colocation providers in particular and the bill’s potential effect on the state’s economy, according to Las Vegas Review-Journal. Economic development officials and a top economist are for it, saying tax breaks would have a positive effect on the state economy. They cited Switch’s Las Vegas and Reno investments as examples of the benefits colo facilities bring. In a Tuesday hearing on the bill, Jeremy Aguero, principal of Applied Analysis, said the numbers that underlie this element of Nevada’s economy are nothing short of massive. In Southern Nevada, the Switch expansion would equate to $5 billion in assets. Aguero said the measure would create thousands of jobs, generate hundreds of millions of dollars in salaries, as well as hundreds of millions in tax revenue, according to the Review-Journal report. In North Dakota, local news site Inforum reported the bill passed the House by a vote of 64 to 27. It was amended to limit exemption to the first four data centers certified by the Tax Commissioner’s office. The tax break would apply to new and refurbished data centers that meet certain criteria. North Dakota is completely outside of natural disaster heat maps and has a few enterprise data centers, but no colocation. The bill is an attempt to spark some initial activity. (Comment on this)
6:42p	Google Compute Engine, AOL Mail Suffer Early Morning Outages Today started with a pair of outages. Google Compute Engine suffered downtime across multiple zones lasting roughly an hour, while AOL Mail suffered an extended outage most of the morning. Google Compute Engine Outage A network issue around midnight Pacific caused loss of connectivity to multiple zones, with the company updating users on the cloud outage roughly every half hour. Google said problems were resolved shortly after 1 a.m. A detailed analysis is pending following internal investigation, with network issues blamed as the culprit. The cloud outage was technically during off-peak hours, but in today’s global connected economy, there’s no such thing as “off peak.” Google Compute Engine suffered problems with its load balancing service for 3 hours in January, with customers unable to create forwarding rules. Google’s Infrastructure-as-a-Service cloud had a total of 4.5 hours of downtime last year across more than 70 outages, according to CloudHarmony. The outage this morning was on the long side. CloudHarmony said Amazon Web Services was the most reliable IaaS cloud last year, with 2.4 hours of downtime total. AOL Mail Down AOL’s email service suffered a widespread outage beginning around 4 a.m. Eastern. The outage appears to have been resolved, although some customers still report ongoing issues. While AOL is past its heyday in the 90s, a big discussion flared up in 2011 about an AOL email address acting as a status symbol. The unfashionable domain became temporarily fashionable, with a certain prestige attached. Part of it was identifying executives and other big shots still using AOL email addresses as early adopters with nothing to prove to anyone. Part of it was ironic usage. AOL has not disclosed the cause of the problem yet, other than claiming there was a network issue. (Comment on this)
11:03p	Cloud Hadoop Distro Microsoft’s First Azure Service Running on Linux Microsoft rolled out a version of HDInsight – its cloud Hadoop distribution – that runs on Ubuntu at this week’s Strata + Hadoop World Conference in San Jose, California. While Linux instances have been available on Azure Infrastructure-as-a-Service cloud, this is the first Microsoft-managed and Azure-hosted service that runs on Linux. Most Hadoop deployments run on Linux, and Microsoft’s move puts HDInsight in a better position to attract more developers with Hadoop experience. The company has worked on making Hadoop more Windows-friendly since at least 2012, when it first submitted a proposal to improve Hadoop for Windows Server and Azure to Apache. Hortonworks, a leading enterprise Hadoop distribution vendor, has also been involved in that effort. One potentially big new capability is extending Linux-based Hadoop clusters sitting in customers’ own data centers to Azure and using the same Linux tools, documentation, and templates across both environments. The service is currently available in four Azure regions – two in U.S., one in Europe, and one in Asia – but the company plans to launch it in all regions over the next several months. Microsoft chose Ubuntu, the popular enterprise Linux distribution by Canonical, because it was “the leading scale-out Linux,” T.K. Rengarajan, corporate vice president of Data Platform at Microsoft, wrote in a blog post announcing the news. Canonical will also provide support for HDInsight on Azure, according to a separate blog post by Corey Sanders, director of program management at Azure. “Customers can leverage the same skills and tools they use for their Linux Hadoop deployments on premises and migrate to Azure or create simple hybrid deployments without having to manage the infrastructure,” Sanders wrote. The Azure cloud VM service supports six Linux distributions. According to Sanders, one out of five customer VMs deployed on Azure run on Linux. Microsoft has been steadily changing its old reputation of a stuffy software giant opposed to all things open source. Besides being more relaxed about Linux, the company recently open sourced .NET, the popular developer framework for building Windows applications. Azure supports Docker, the popular open source application container technology that runs on Linux. Microsoft has also created a Windows command line interface for Docker. (Comment on this)

<< Previous Day

2015/02/19 [Calendar]

Next Day >>