Uptime: Colocation Firms are Building Fewer Data Centers

June 24, 2016 by Yevgeniy Sverdlik Posted in Industry Insights & News

Brought to you by Data Center Knowledge

If you look at recent earnings reports by the biggest data center providers, you’ll get the impression that the industry is booming.

And it is. Enterprises are moving more workloads either to the cloud or to commercial colocation facilities, and data center providers are benefiting from both. As more companies use cloud services, cloud providers are racing to lease as much data center capacity as they can get their hands on, resulting in a boom for the big data center providers who can’t build new facilities fast enough to satisfy all the demand.

The sound of champagne corks popping after earnings reports by the biggest players in the market, however, can mask the fact that in general, the amount of new data centers being built for lease by one or multiple tenants in the US has been declining.

This doesn’t necessarily mean the amount of new data center capacity being brought to market is shrinking. This is the age of the mega data center: service providers may be building fewer facilities, but the size of each individual building is getting bigger and bigger.

Market studies by IDC in 2014 and 2015 found that the trend in the data center provider industry was toward building fewer but larger buildings.

In a more recent study, 24 percent of colocation providers Uptime Institute surveyed this past February said their company had built a new data center within the previous 12 months. That’s down from 29 percent in 2015 and 45 percent in 2014.

Interestingly, construction slowdown in the data center provider industry has been more drastic than the slowdown in enterprise data center construction. Fifteen percent of enterprise IT respondents said their company had built a new data center within the previous 12 months both this February and the year before. Eighteen percent said so in 2014.

Source: Uptime Institute Data Center Industry Survey 2016

Far from Perfect

While overall colocation customer satisfaction levels are high – only 7 percent of respondents to Uptime’s survey said they were dissatisfied or very dissatisfied with their primary data center provider – colocation isn’t the perfect answer for everybody. According to Uptime, 40 percent of enterprise IT respondents were paying more for colo contracts than they expected to pay when they signed those contracts.

Nearly one-third said they had experienced a data center outage at a colocation site, and the bulk of enterprise respondents said downtime compensation in their agreements with colo providers was insufficient. About 60 percent said the cost of data center outages overshadowed whatever downtime penalties were included in their Service Level Agreements.

Lots of Business Still on the Table

While many of the biggest data center providers are chasing the multi-megawatt wholesale deals with cloud giants, there is a huge portion of the enterprise market that remains untapped, and companies like Equinix, QTS Realty, and CyrusOne, as well as the cloud giants themselves, are pursuing that opportunity.

Source: Uptime Institute Data Center Industry Survey 2016

Enterprise-owned data centers still host 71 percent of enterprise IT assets, according to Uptime. Data center providers have 20 percent of those assets, while the remaining 9 percent is in the cloud.

The big question today is how much of that 71 percent will go to the cloud, and how much of it will end up in colocation data centers.

Further reading: Why Keep the Enterprise Data Center?

Source: TheWHIR

Trusted Advisors: Partners for Success

June 24, 2016 by Theresa Caragol Posted in Industry Insights & News

Trusted Advisors: Partners for Success

The notion that CEOs and Executives are the sole visionaries who can lead and drive a company to success with no outside help is long gone. Leaders who leverage outside help go farther faster and research shows they are significantly more successful in both the short and long term. Whether you are a CEO, a C Suite Executive, or an emerging and/or aspiring leader, there are several core considerations as we contemplate our best partner advisors to assist leaders and corresponding organizations with achieving success.

Below are some common challenges that advisors often consult with CEOs and business executives about:

Strategy around what markets my company wants to be in, globalization and who will carry my brand
Differentiation and staying ahead of competition and/or market challenges like overcoming obsolescence
Effectively retaining and engaging employees and people who currently or in the future could work with my company
Alliances formed around my business or partners who will take my company’s solutions and services to market
Verticals / Clients / partners I sell to – leads, sales, customer loyalty
Product lines and offers, directions, services
Financial, tax, and operational challenges – managing finances tactically and strategically
Capital optimization – what’s needed for short and long term success
IP and legal challenges
Unique industry specific challenges
Growth as a leader
The Executive Brand I need to represent my business

Debbie Tyler, accomplished CEO, and Vistage Chair coaches business leaders and shared some of her perspective and guidelines for assembling a first class team of advisor partners.

Guideline #1: SKILLS & EXPERIENCE – Ensure your team of advisors cover relevant and diverse skills and experience encompassing financial, technical, people and operations, the business outlook, and relationships with your customer and prospects.

Guideline #2: TALENT – One of the top issues of executives today centers around talent. How does an organization affordably find the right talent, hire the right employees, challenge them, keep and protect them? Having a set of advisors with experience in successful corporate culture, human capital, incentive approaches and employee engagement, could help look at this subject from a number of viewpoints. Experienced advisors give valuable insights; likely resulting in different solutions that could positively impact your situation.

Guideline #3: MIX – Consider the mix of advisors: having expertise from diverse backgrounds, diverse age groups, and diverse types of experience is important. Statistics support this. In a recent blog http://www.theresacaragol.com/blog/, we uncovered that profitability, innovation, and diversity are directly correlated according to extensive global 2016 research.

Evaluate potential advisors outside of your network and don’t always rely on your most trusted relationships. It’s potentially dangerous to select advisors solely based on proximity or comfort, especially relationships of folks who are vested in you. It’s important that advisors are people who will give unfiltered, sometimes provocative, innovative or even controversial insights. It’s often in this type of feedback that golden nuggets for success exist.

Guideline #4: COMPENSATION – Consider the compensation you give your advisors. Is it an informal mentor or sponsor relationship that has no compensation? Is it a relationship that carries an equity or dividend stake in the business associated with company growth? Is it a standard advisory board that has a set stipend compensation? Your approach is dependent upon the stage of the company, the purpose of the advisors, and the sophistication and experience of the executive seeking the advice, among other factors. Most importantly, take care to properly set up and formalize the relationship with the advisor. Be clear with all advisors about expectations, time and resource commitments, compensation model, and the timeline of the advisory role.

Guideline #5: EMOTIONAL INTELLIGENCE – Test your own emotional intelligence; ensure you have a high degree of self-actualization and are brave enough to receive the input, as well as see and work on your own blind spots. Are you truly open to hearing and receiving feedback and taking action on it? High performing executives can handle confrontation and challenges from those around them

Debbie Tyler asserts the right group of advisor should challenge you via healthy debate. Top executives are assertive and confident but comfortable and frequently energized when challenged by their advisors. If you don’t receive questions you haven’t thought of already, you likely are missing out on opportunity – money, footprint, or some area for growth.

Last, Debbie Tyler offers another key tip for assembling your board. Consider an empty room and think about where it is you need to take your organization for success. Then list out the types of skills, experience, personalities, and connections required at the table to help you scale your mountain. Then and only then do you place a name; unless of course it’s a no brainer like Warren Buffet!

How many advisors and what types of boards an individual should participate in is dependent upon the complexity of the role the person holds, the cross-functional nature of what his/her mission is, and the degree of change required for the organization’s success. In general, you may choose five to eight individuals with different backgrounds who can offer strong consult and perspective.

Other ways to obtain advisory services that can be extremely effective include: formal CEO and executive peer leadership groups like Vistage that offer advisory, peer networking, coaching, and global network access all in one. Another option is to hire an individual executive coach. According to a recent Fast Company article, 60 percent of growth stage CEOs use executive coaches, and 32 percent of seed-stage CEOs use a coach.

There are numerous ways to get trusted and confidential perspective, to learn from other seasoned and fellow executives and to continuously invigorate and challenge your thinking as an executive. Be deliberate about the advisors, be dedicated to listening and gaining consult, and be diverse in the group you assemble.

These factors will be a key tenet of you and your organization’s success.

This article is brought to you by HostingCon, the Cloud and Service Provider Ecosystem event. Join us in New Orleans, Louisiana July 24-27, 2016 to hear Theresa and other thought leaders talk about issues and trends in the cloud, hosting and service provider ecosystem.

Save $100 off your HostingCon All Access Pass with coupon code: H1279

Source: TheWHIR

Why the UK's vote to leave the EU will have little effect on its data protection rules

June 24, 2016 by Peter Sayer Posted in Industry Insights & News

Why the UK's vote to leave the EU will have little effect on its data protection rules

With the haircut that the sterling-euro exchange rate has taken in the wake of the U.K.’s vote to leave the European Union, the U.K. has suddenly become a low-cost country for companies wishing to host or process the personal information of EU citizens.

EU businesses will need to weigh that price cut against the regulatory uncertainty Thursday’s vote introduced — but it turns out that’s surprisingly small, at least in the short to medium term.

As for U.K. businesses hoping for more relaxed data protection rules in the wake of the referendum vote, they will have to wait — perhaps for a very long while.

That’s because many of the rules that the 51.9 percent who voted to leave the EU hoped to escape are, in fact, firmly part of U.K. law, and will only go away if the U.K. parliament votes to repeal them.

And it can’t do that until it has negotiated its exit from the EU, which is a matter of international treaty and not the will of the people.

The first question, then, is when will the U.K. officially leave the EU?

That will depend on when the U.K. government informs the other member states of its intention to leave by invoking Article 50 of the Lisbon Treaty. The UK will cease to be bound by the EU treaties two years after that date — sooner in the unlikely event that all parties reach an agreement on an exit settlement before then.

However, U.K. Prime Minister David Cameron is in no hurry to invoke Article 50. On Friday morning he announced that he will resign and make way for a new leader of the ruling Conservative Party before the party’s annual conference in October. Invoking Article 50, he said, would be a task for his successor.

That means the U.K. is likely to remain part of the EU until October 2018 — or longer, if Cameron’s successor is in no rush to invoke Article 50.

That means U.K. businesses and citizens will still be subject to EU laws for some years to come.

Those laws come in two forms: directives, and regulations. In the field of data protection, there’s one of each to pay attention to.

The most significant — for now — is the 1995 Data Protection Directive.

Directives are proposed by the European Commission (the members of which are nominated by the EU member states), then amended by the European Council (composed of the heads of the EU member governments or their ministers) and the European Parliament (directly elected by EU citizens) until all three parties reach a compromise. Then, the parliaments of each member state transpose the directives into their own national law, adapting it where necessary to fit their own legal systems and circumstances. In this way, the Data Protection Directive took effect in 1998.

One of its key provisions, for businesses at least, is that EU citizens’ personal information may only be processed in countries offering a level of data protection at least equal to that afforded by EU law.

Since the U.K.’s data protection regime will remain unchanged, for now, U.K. businesses can still process data for EU companies and citizens, and U.K. citizens will have the same protections if their data is exported to, say, the U.S.

Protection of EU citizens’ data in the U.S. has itself been called into question since the October 2015 decision by the Court of Justice of the EU to overturn the legal instrument providing that protection, the so-called Safe Harbor Agreement. EU and U.S. officials are still negotiating the details of its replacement, Privacy Shield, which will also cover the U.K. until it formally leaves the EU.

The other EU data protection law of relevance to the U.K. is the General Data Protection Regulation (GDPR), voted in April 2016. This introduces harsher fines for companies breaching the rules — up to 4 percent of worldwide revenue — and seeks to harmonize those rules, eliminating national differences allowed under the Data Protection Directive.

Regulations begin life in the same way as directives, as compromise texts agreed upon by the Commission, Council and Parliament. After that, though, there’s no time-consuming transposition into national laws: Regulations are directly applicable, and automatically enter effect after two years.

At first sight, that would suggest that U.K. citizens will benefit from, and U.K. businesses will be subject to, the effects of the GDPR from April 2018 through at least October 2018.

That, though, is without considering the exemptions from EU home affairs and justice legislation negotiated by the U.K., Ireland and Denmark. The exemptions mean the GDPR will apply only partially in the U.K up until October 2018.

But what then? Well, one of the innovations of the GDPR is that the rules applicable depend on the location of the data subject, so companies in the U.K. will still have to comply with it when processing EU citizens’ data.

U.K. businesses might even choose voluntarily to follow EU data protection rules at all times, in order to hang on to their U.K. customers.

“It would make no sense at all for U.K. regulations to be any less stringent. Poor safeguards against loss, theft and misuse of data would ultimately cost U.K. business, as consumers and brands put their data elsewhere,” said Richard Lack, EMEA director of sales at Gigya, which provides a visitor tracking and identification service for websites.

Following the EU data protection rules would be a good thing for U.K. businesses in other respects, according to Javvad Malik, security advocate at AlienVault, a security threat management company.

“Many Infosec professionals seem to view the legislation in a positive light, believing that stipulations such as ‘data protection by design’ will make the data held by their organizations more secure,” he said of the GDPR.

Until October 2018, then, and even beyond, it seems unlikely that much will change, in the field of data protection at least.

Source: InfoWorld Big Data

United Internet Said to Weigh Bid for Cinven's Host Europe

June 23, 2016 by Bloomberg Posted in Industry Insights & News

United Internet Said to Weigh Bid for Cinven's Host Europe

By Stefan Nicola and Manuel Baigorri

(Bloomberg) — United Internet AG is considering a bid for Host Europe Group Ltd., the internet domain registry and web-hosting company owned by Cinven Ltd., according to two people with knowledge of the information.

United Internet, which offers web hosting along with internet access and website-making tools, is working with advisers to evaluate a potential bid, said the people, who asked not to be named discussing plans that haven’t been announced. The Montabaur, Germany-based company hasn’t decided whether to pursue an offer and may not do so, one of the people said.

A purchase of Host Europe would bolster an increasingly important business of web hosting and cloud computing at United Internet, as more consumers store data online and companies hook factories to the web. United Internet is open to buying international web-hosting companies with a focus on its existing markets, Chief Executive Officer Ralph Dommermuth said in an interview last month.

United Internet declined to comment. Cinven had no immediate comment.

Deutsche Telekom AG, which owns the Strato AG web-hosting business, is also considering a bid for HEG, a person familiar with the situation said last week.

United Internet rose 1.3 percent to 39.15 euros at 14:47 p.m. Frankfurt. Drillisch AG fell as much as 6.4 percent in Frankfurt after the news, which quelled speculation that United Internet may raise its stake in the wireless operator, according to Jochen Reichert, an analyst at Warburg Research.

“A potential bid for HEG would not leave room for a takeover of Drillisch at the same time,” Reichert said Thursday.

In past years United Internet has acquired web-hosting companies including Arsys Internet and United-Domains AG. It has been weighing an initial public offering of its business applications division, which competes with domain registration services and web-hosters including GoDaddy Inc.

Cinven, a private equity firm based in Guernsey, Channel Islands, bought Host Europe from Montagu Private Equity LLP for 438 million pounds ($651 million) in 2013, expanded the business with acquisitions including of German hosting company Intergenia Holding GmbH last year.

The transaction value of Host Europe is 584 million pounds including “aggregate buy and build” values, according to Cinven’s website.

Source: TheWHIR

Equinix Extends Microsoft Azure ExpressRoute Availability Into Canada Azure

June 23, 2016 by admin Posted in Industry Insights & News

Equinix Extends Microsoft Azure ExpressRoute Availability Into Canada Azure

Equinix, Inc. has announced that it will provide private access to Microsoft Azure ExpressRoute via its Equinix Cloud Exchange™ in the company’s Toronto (TR2) International Business Exchange™ (IBX®) data center. Through this partnership, Canadian-based companies gain private access to both Azure and Office 365 while maintaining data residency in region. This deployment is part of Equinix’s global partnership with Microsoft which now directly connects 18 markets globally with Azure ExpressRoute.

“We are thrilled to continue collaborating with Equinix and are happy to welcome them in Canada as an Azure ExpressRoute partner. Equinix now connects to 18 markets globally with Azure ExpressRoute, helping to further business innovation and competitiveness in a cloud-first, mobile-first world,” said Janet Kennedy, president, Microsoft Canada.

Many companies are looking to deploy hybrid clouds through private interconnection, to reap the scalability and cost benefits of the cloud, while at the same time maintaining the control of on-premises infrastructure. In fact, a recent study, Enterprise of the Future, found that by 2017, 84% of IT leaders will deploy IT infrastructure where interconnection — defined as private, secure physical or virtual connections — is at the core, compared to only 38% today. By offering private access to Azure ExpressRoute and Office 365 across the global market, Equinix helps enterprise CIOs achieve this by seamlessly incorporating Azure and Office 365 services into their existing architectures.

Opened in 2015, Equinix’s TR2 facility offers interconnections for more than 155 enterprise companies, financial services firms, and more than 60 cloud and network providers that have collocated with Equinix via the company’s original data center in Toronto, TR1. Expanding Azure ExpressRoute into this critical market provides our many Canadian enterprise customers the benefit of a high-performing and private connection to the Microsoft Cloud and our growing cloud ecosystem.

Equinix is working with several reseller partners in the Canadian market to help customers quickly realize the benefits of Equinix Cloud Exchange and the Microsoft Cloud. These partners include Airgate, IMP Solutions, New Signature, Powerland and Solgenia.

“By providing direct access to Azure and Office 365 via ExpressRoute inside Equinix data centers, we are enabling companies, worldwide, to bridge their cloud and data center strategies and fully realize the benefits of hybrid and multi-cloud. By offering this capability now in Canada, local companies concerned with data residency can now get the full benefits of Microsoft cloud services, in a trusted, secure, low-latency deployment,” said Greg Adgate, vice president of global technology partners, Equinix.

Source: CloudStrategyMag

BlueData Introduces BDaaS For On-Premises And Cloud Deployments

June 23, 2016 by admin Posted in Industry Insights & News

BlueData Introduces BDaaS For On-Premises And Cloud Deployments

BlueData has announced that the enterprise edition of its BlueData EPIC software will run on Amazon Web Services (AWS) and other public clouds. The BlueData EPIC enterprise edition is now in directed availability for AWS. General availability for AWS — as well as availability for Microsoft Azure, Google Cloud Platform, and other public cloud services — will be released in the coming months. Customers will have the same BlueData user experience, distribution flexibility, and application configurability for Hadoop and Spark deployments on-premises and/or in the cloud.

BlueData’s software innovations make it easy to deploy Big Data infrastructure and applications, leveraging embedded Docker container technology. The BlueData EPIC software platform provides a self-service Big-Data-as-a-Service (BDaaS) experience with the highest levels of security and performance for Big Data analytics in the enterprise. BlueData has offered a free community edition of BlueData EPIC running on AWS since last year; but until now, BlueData’s enterprise edition was available only for on-premises deployments.

BlueData delivers simplicity, agility, flexibility, and cost savings for Big Data deployments — now in both on-premises and cloud-based environments. Data scientists, analysts, and data engineers no longer have to deal with the complexities of their underlying infrastructure or cloud provider. They can choose their own distributions and applications for Big Data analytics, and tap into data either on-site or in the cloud. IT teams can control costs with resource quotas, and rely on BlueData’s multi-tenant security and governance capabilities to meet enterprise-class auditing and regulatory compliance requirements.

“Big Data has been too darn hard to implement. Cloud deployments provide a simpler path and are helping to fuel the next wave of Big Data adoption,” said Tony Baer, principal analyst for Big Data at Ovum. “BlueData is well positioned to help enterprise customers on this next phase of their Big Data journey, whether on-premises or in the cloud — or more likely, some combination of the two.”

The benefits of the BlueData EPIC software platform for cloud deployments include:

Consistent user experience: BlueData will offer the only unified Big-Data-as-a-Service software platform for hybrid (on-premises and cloud) and multi-cloud (e.g. AWS, Microsoft Azure, Google Cloud Platform) deployments. End users and administrators will have an easy-to-use “single pane of glass” for creating and managing Big Data environments; and the Docker images for Hadoop, Spark, and other Big Data applications will be the same irrespective of the underlying infrastructure or cloud service.

Greater flexibility and choice: Only BlueData provides customers with the ability to create their own unique BDaaS environments — with their preferred Big Data platforms (and versions) and applications, either commercial or open source. They can have multiple environments for different use cases, for different user groups, and for dev/test/QA or production. And they’ll have the flexibility to update their Big Data distributions and applications on their timeline, not the timeline of the cloud service provider.

Eliminate data movement: BlueData will provide the first and only solution that enables customers to tap into both cloud-based data stores (such as Amazon S3) as well as on-premises data for analytics with Hadoop and Spark. This allows them to leverage the elastic compute of the public cloud while keeping data on-premises — avoiding the cost and complexity of duplicating, copying, and transferring data to the cloud service.

“The BlueData software platform is the first Big-Data-as-a-Service solution available for both on-premises and public cloud environments,” said Kumar Sreekanti, CEO of BlueData. “The future of Big Data analytics will be neither 100% on-premises nor 100% in the cloud. We’re seeing more multi-cloud and hybrid deployments, with data both on-prem and in the cloud. BlueData provides the only solution that can meet the realities of these mixed environments in the enterprise.”

Source: CloudStrategyMag

Lightower Cloud Connect Enables All-Fiber, Direct Paths To Major Cloud Providers

June 23, 2016 by admin Posted in Industry Insights & News

Lightower Cloud Connect Enables All-Fiber, Direct Paths To Major Cloud Providers

Lightower Fiber Networks has announced that its Cloud Connect solutions enable end-to-end fiber connectivity directly to major cloud providers, including Amazon Web Services and Microsoft Azure. Lightower Cloud Connect offers customers a wide range of flexible bandwidth and service options including Ethernet, wavelengths, and dedicated Internet access, with bandwidths up to 10 Gbps.

As customers continue to move vital business functions to the cloud, the network connecting the organization to their cloud services has become more critical than ever. With over 30,000 route miles of fiber serving over 15,000 service locations in 17 states throughout the Midwest, Northeast, and Mid-Atlantic, Lightower can deliver end-to-end, fiber connectivity between organizations and the dedicated on-ramps to cloud providers. An all-fiber network between organizations and the cloud enables the network performance needed to ensure cloud-based applications run smoothly, data and files transfer quickly, and latency can be minimized.

The Cloud Doesn’t Work Without the Network (SM)

Lightower offers direct, fiber connectivity to Amazon Web Services at the following locations:

32 Avenue of the Americas, New York City, NY

21715 Filigree Court, Ashburn, VA

Ethernet, wavelength, or Internet-based connectivity

Bandwidth options from 50 Mbps to 10 Gbps

Lightower also offers direct, fiber connectivity to Microsoft Azure via the Equinix Cloud Exchange at the following locations:

60 Hudson Street, New York City, NY

21715 Filigree Court, Ashburn, VA

Ethernet, wavelength, or Internet-based connectivity

Bandwidth options range from 100 Mbps to 10 Gbps

“The cloud is becoming an essential resource for more and more organizations of all types. Businesses, governments, and educational organizations are all adopting cloud platforms for applications that require very high levels of network performance,” explains Doug Dalissandro, chief revenue officer, Lightower. “Lightower Cloud Connect offers superior network performance for all types of cloud applications. For added resiliency, Cloud Connect solutions can be designed with redundancy and diversity. Of course, all of this is backed up by Lightower’s industry-leading customer support.”

Source: CloudStrategyMag

HDFS: Big data analytics' weakest link

June 23, 2016 by Andrew C Oliver Posted in Industry Insights & News

HDFS: Big data analytics' weakest link

For large-scale analytics, a distributed file system is kind of important. Even if you’re using Spark you need to pull a lot of data into memory very quickly. Having a file system that supports high burst rates — up to network saturation — is a good thing. However, Hadoop’s eponymous file system (Hadoop Distributed File System, aka HDFS) may not be all it’s cracked up to be.

What is a distributed file system? Think of your normal file system, which stores files in blocks. It has some way of noting where on the physical disk a block starts and how that block matches to a file. (One implementation is a file allocation table or FAT of sorts.) In a distributed file system, the blocks are “distributed” among disks attached to multiple computers. Additionally, like RAID or most SAN systems, the blocks are duplicated so that if a node is lost from the network then no data is lost.

What’s wrong with HDFS?

In HDFS, the role of the “file allocation table” is taken by the namenode. You can have more than one namenode (for redundancy), but essentially the namenode constitutes both a failure point and a type of bottleneck. While a namenode can fail over, that does take time. It also means keeping in sequence, which introduces more latency. In HDFS there is also some threading and locking stuff that happens as well as the fact that it is garbage-collected Java. Garbage collection — especially Java garbage collection — requires a lot of memory (generally at least 10x to be as efficient as native memory).

Moreover, in developing applications for distributed computing we often figure that whatever inefficiency we inject in language choice will be outweighed by I/O. Meaning so what if it took me 1,000 operations to open a file and give you some data, because the time it took for an I/O operation was 10x that. Simplistically speaking, the higher level the language, the more operations or “work” is executed per line of code.

That said, the lower level the component, the more we pay for this inefficiency. Meaning it is rational to trade lower operation performance for shorter time to develop and deploy, lower maintenance costs, and better security (i.e., buffer overruns and underflows are next to impossible in a high-level, garbage-collected language like Java). However, as you go lower level the inefficiency catches up. This is why most operating systems are written in C and Assembly as opposed to, say, Java (despite several attempts). One could argue that a file system is at that lower level.

The tooling around HDFS is kind of poor compared to any file system or distributed store you’ve ever dealt with. You’re asking your IT operations people to administer a Java-based file system that, at best, implements bastardized versions of their “favorite” POSIX tools. Yes you can mount HDFS with NFS, but have you actually tried that? How well did that work out for you, really? The other tools for mounting HDFS are also pretty poor. Instead, you deal with weird REST bridge tools and a command-line client that doesn’t even accept most options for ls, let alone anything else.

There are of course other particular aspects of HDFS that are inefficient or just problematic. Most derive from the fact that HDFS is a file system written in Java, of all things.

What about HDFS could be fixed? HDFS has native code extensions to make it more efficient. Meanwhile, the community has improved the namenode substantially. However, on higher-end systems with a lot of operations you still hit a namenode bottleneck that you can see in your favorite monitoring and diagnostic tools. Moreover, kicking the namenode over to solve ghost problems is something that occasionally has to happen. Overall, a more mature distributed file system written in C or C++ with mature bindings to common operating systems is often a better option.

Spark and cloud demand change

Many of the early corporate Hadoop deployments were done on-premises, and the sales compensation plan for the early vendors was based on this assumption. With the rise of Spark and cloud deployments, it isn’t uncommon to see Amazon S3 used as a data source.

The Hadoop vendors all had a vision of a more unified Hadoop platform where HDFS would integrate with security components (Cloudera and Hortonworks of course doing their own separate and incompatible security systems). However, with MapReduce giving way to Spark, and Spark being a lot more ambivalent about the type of file system, the “tight” integration with the file system seems less critical let alone feasible.

Meanwhile, alternative file systems such as MapR’s MapR FS are gaining in interest as companies discover the joys of dealing with HDFS in an operational setting. Moreover, the design of MapR FS doesn’t include namenodes, but uses a more standard and familiar cluster master election scheme. MapR’s partitioning design should also work better to avoid bottlenecking.

There are other options like Ceph, or an object store like S3, or a more standard distributed file system like Gluster. I/O Performance tests [PDF] tend to yield very positive results for Gluster. If you’re mainly looking to store files that you’re going to suck into Spark, there are rational reasons to pick something that is more operationally familiar to your IT operations folks and yet is, well, faster.

I’m not saying that large HDFS installations are going to migrate overnight, but that as we see more Spark and cloud projects we’re likely to see less and less HDFS over time. That will leave YARN as the remaining piece of Hadoop that people will still use. Maybe.

Source: InfoWorld Big Data

Cloud or on-prem? This big-data service now swings both ways

June 22, 2016 by Katherine Noyes Posted in Industry Insights & News

Cloud or on-prem? This big-data service now swings both ways

There are countless “as-a-service” offerings on the market today, and typically they live in the cloud. Back in 2014, startup BlueData blazed a different trail by launching its EPIC Enterprise big-data-as-a-service offering on-premises instead.

On Wednesday, BlueData announced that the software can now run on Amazon Web Services (AWS) and other public clouds, making it the first BDaaS platform to work both ways, the company says.

“The future of big data analytics will be neither 100 percent on-premises nor 100 percent in the cloud,” said Kumar Sreekanti, CEO of BlueData. “We’re seeing more multicloud and hybrid deployments, with data both on-prem and in the cloud. BlueData provides the only solution that can meet the realities of these mixed environments in the enterprise.”

BlueData’s EPIC (short for “Elastic Private Instant Clusters”) platform taps embedded Docker container technology to let businesses spin up virtual Hadoop or Spark clusters within minutes on their existing infrastructure, the company says, giving data scientists on-demand access to the applications, data, and infrastructure.

BlueData has offered a free community edition of BlueData EPIC running on AWS since last year, but until now the enterprise edition was available only for on-premises deployments.

With the new flexibility to run in hybrid and multicloud environments, customers can eliminate data movement by keeping data on-premises while running compute in the cloud. At the same time, there’s a “single pane of glass” for creating and managing big-data environments, BlueData says. The Docker images for Hadoop, Spark, and other big data applications will be the same regardless of the underlying infrastructure or cloud service.

As of Wednesday there’s limited availability on AWS, with general availability there — as well as on Microsoft Azure, Google Cloud Platform, and other public cloud services — due in the coming months. Pricing on the public cloud will include usage-based and annual subscription options.

Also on Wednesday, BlueData introduced the summer release of its EPIC platform.

Tech jobs report: Security, devops, and big data stay hot

June 22, 2016 by Paul Krill Posted in Industry Insights & News

Tech jobs report: Security, devops, and big data stay hot

If you’re wondering what IT skill sets to acquire, security and devops are doing well in the job market. Pay for cloud skills, however, is eroding.

Research firm Foote Partners’ latest quarterly IT Skills and Certifications Pay Index determined that the market value for 404 of the 450 IT certifications it tracks had increased for 12 consecutive quarters. Market values rose for noncertified IT skills for the fifth consecutive quarter.

Foote’s report is based on data provided by 2,845 North American private and public sector employers, with data compiled from January to April 1. (Noncertified skills include skills that are in demand but for which there is no official certification, Foote spokesman Ted Lane noted.)

Security skills command increasing salaries, with no end in sight

In the security space, Foote found that values for 76 certifications have been on a slow and steady path upward for two years, with an 8.7 percent average increase. The certifications’ values have risen 6.3 percent in the past year. “Strong-performing certifications in the first three months of 2016 were those in IT security management and architecture, penetration testing, forensics, and cybersecurity,” the report said.

Security needs will only grow as cybercrime and hacking continue to grow, Foote said. “Employers are becoming much more aware that they don’t have the right people in their security,” said David Foote, chief analyst at Foote Partners.

Devops demand gains serious traction

The report also finds devops gaining “serious traction,” with companies looking to bridge the gap between developers and operations. The latest pay premium data for 2,745 employers shows a gain of 14.3 percent in average market value for devops skills in the past 12 months, with devops engineers in big demand. Foote’s most recent data shows that devops engineers’ salaries averaged $119,709, while lead devops engineers’ salaries averaged $134,512.

Big data demand is for skills more than certifications

Meanwhile, the need for big data skills also continues to lead to pay increases — about 8 percent over the last year. “Noncertified skills were the big winners, with 47 big-data-related skills rising 6.7 percent in market value during this period, while 32 big data certifications were … down 3.5 percent in market value,” the report said.

Cloud skills are worth less as supply catches up to demand

On the flip side, Foote found that although IT organizations continue to invest in cloud services, pay for these skills has started to lose ground. Eighty cloud-related skills (both certified and noncertified) lost nearly 1 percent in pay premium market value between January 1 and April 1, and market values for cloud certifications have lagged behind those for noncertified sills.

Bare Metal Servers and Cloud Server Hosting

Yearly Archives: 2016
Home / 2016 (Page 30)

Uptime: Colocation Firms are Building Fewer Data Centers

Far from Perfect

Lots of Business Still on the Table

Trusted Advisors: Partners for Success

Why the UK's vote to leave the EU will have little effect on its data protection rules

United Internet Said to Weigh Bid for Cinven's Host Europe

Equinix Extends Microsoft Azure ExpressRoute Availability Into Canada Azure

BlueData Introduces BDaaS For On-Premises And Cloud Deployments

Lightower Cloud Connect Enables All-Fiber, Direct Paths To Major Cloud Providers

HDFS: Big data analytics' weakest link

What’s wrong with HDFS?

Cloud or on-prem? This big-data service now swings both ways

Tech jobs report: Security, devops, and big data stay hot

Security skills command increasing salaries, with no end in sight