CDNs using OnApp’s federation (OnApp also has its own effort here, CDN.net) can tap into the spare capacity of service providers around the world, with more than 150 points of presence (PoPs) theoretically being at their disposal. CDNify is launching with 40 PoPs, and it’s trying to win over customers on price and simplicity, as founder James Mulvany told me:

“We want to have a very low cost of entry, starting with a free account – you can start using it straight away, to play with the system. We’ve got things like nice reporting, good graphs so you can see your usage, decent support and video tutorials.

“The front-end system that customers will use is something we’ve built in-house. It’s fairly simple, although we’ve got lots of exciting features that we’ll be rolling out over the next year. We’re trying to create a very clean experience.”

In terms of price, CDNify charges $0.05/GB/month, which significantly undercuts both Amazon and Rackspace ($0.12/GB/month for the first 10TB of traffic). Additionally, Mulvany said, “Amazon charges you for the number of hits you get, whereas we just charge on bandwidth.”

CDNify is privately funded off the back of the founders’ other business, internet radio outfit Wavestreaming — and, as it happens, the launch of that service provided the impetus for CDNify. “We started looking at CDNs and found they were aiming towards the enterprise, the big guys,” Mulvany said. “We decided to build CDNify for people like us: web companies, startups and mobile app developers.”

Mulvany said CDNify expects its initial growth to come from the North American and European markets, although the company is “not targeting any specific region.”

Here’s a video pitch:

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Federated clouds: for when one cloud isn’t good enough
• Infrastructure Q1: Cloud and big data woo enterprises
• A near-term outlook for big data

While public cloud services, including content delivery networks (CDNs), are usually very reliable, no one is perfect. Outages happen, and as a result some companies find themselves looking into multi-cloud strategies to ensure redundancy (and to optimize performance and cost). The issue is that cloud costs and resource allocation are complex — hence the emergence of a new breed of cloud mediation services such as Rightscale and Cedexis.

DT is preparing two complementary services in this space. The first is NetAnalyze, which draws on the billion network measurements that DT’s “community” takes every day, spanning 32,000 networks in 230 countries. Webmasters can put the NetAnalyze tag on their site and visiting customers will then automatically generate anonymized measurement for metrics like throughput and response time.

Then NetOptimize kicks in. When a customer requests the website from wherever they are located, NetOptimize will use the NetAnalyze metrics to determine which provider will deliver the content most quickly, and automatically route the content accordingly. Pricing for this load-balancing service is pay-per-use. The result, in theory, is better performance and lower risk of outages, and also better price-to-performance ratios, given the ability to hop between different providers according to needs.

DT’s website also touts the fact that such multi-sourcing approaches make it easier to avoid vendor lock-in. The company says NetAnalyze and NetOptimize make it possible to “form a unified strategy across multiple platforms (cloud, data center or CDN)”.

A glance at the NetOptimize portal (which appears to default to Japanese, at least from my end) shows that the service covers numerous clouds and CDNs. On the cloud side, we have locations for Amazon EC2, Google App Engine, GoGrid, InstaCompute, Internap AgileCLOUD, Joyent, PhoenixNAP, Profitbricks, Rackspace Cloud, Softlayer and Windows Azure. For CDN, there’s Akamai, Azure, BitGravity, CacheFly, CDN77, CDNetworks, CDNVideo, ChinaCache, ChinaNetCenter, CloudFlare, Cloudfront, Edgecast, Fastly, Fastweb, Highwinds, Internap AgileCAST, Internode, Level3, Limelight, NetDNA, Ngenix, OnApp, Pacnet, UPX CloudCache and Yacast.

I’ve asked DT for further details of the service, such as when they intend to officially take the wraps off it, and will add their response when I get it.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Migrating media applications to the private cloud: best practices for businesses
• Report: Delivering Content in the Cloud
• 12 tech leaders’ resolutions for 2012

OnApp started out selling cloud orchestration software to hosting providers that wanted to get into the public cloud business. Over time, the firm built a sizeable federation of service provider customers, giving them the ability to use each others’ spare capacity – federated CDN launched in 2011; then came distributed storage; and federated compute capacity is next on the horizon.

However, until now OnApp’s game has been all about helping service providers make the most of their spare resources within the federation – one provider may lack a point of presence (PoP) in a certain location but be able to use that of a fellow federation member, for example. CDN.net is OnApp’s first attempt at selling that federated capacity directly to end users, in this case companies that want to boost the performance of their websites in various locations around the globe.

CDN.net offers access to over 150 PoPs, which is not as extensive a network as those offered by Akamai and Limelight, but way bigger than those from smaller players such as Yottaa, MaxCDN and CacheFly — it’s even slightly more wide-ranging than CDNetworks’ network. However, according to CDN.net marketing manager James Fletcher, the real selling point is CDN.net’s flexibility and pay-by-usage pricing:

“It allows the end user to be in control of what they are purchasing … We saw in the marketplace that you can buy CDN and get a one-size-fits all solution, but that doesn’t work for everyone. The end result is you pay for resources and locations you don’t use. The CDN.net vision allows you to spin up on the fly and provision and customize as you need to.”

At launch, CDN.net will only include 30 PoPs with a focus on Europe, North America and Asia-Pacific. However, OnApp is working to add locations in emerging markets and users will be able to add locations based on demand. “If the customer comes along and wants somewhere in South Africa, we can work with the service provider network to get one up and running,” Fletcher explained. Livestreaming capabilities will also go live soon.

Ultimately, OnApp is trying to “help line the pockets of the service providers”, as Fletcher put it, but it’s also quietly becoming one of Europe’s most significant cloud players, perhaps the most significant. Others have talked about or even attempted this kind of federated model, but no-one has achieved the sort of scale that OnApp can boast – scale that it achieved by stealth, but that it’s now starting to exploit in earnest.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Report: Delivering Content in the Cloud
• Infrastructure Q1: Cloud and big data woo enterprises
• Federated clouds: for when one cloud isn’t good enough

First off, it sliced the cost of its Content Delivery Network (CDN) services by a third from $0.18 to  $0.12 per GB. The new, lower price matches Amazon CloudFront CDN on-demand per-GB pricing for the first 10 TB per month of out-bound data transfer. CloudFront prices drop after that.  Amazon does not publish its reserved pricing for CloudFront. CDNs route traffic around the web to put it closer to prospective users thus reducing latency and page load times.

Secondly, the new tiered price structure will first apply to Cloud Files Object storage with volume discounts ranging from $0.10 per GB per month for up to 1 TB, then falling to $0.075 per month or lower when storage surpasses a petabyte (see chart.)

For comparison, AWS S3 storage prices are lower, but, as Rackspace likes to note, Amazon also charges for PUT, POST, LIST, HEAD, GET and DELETE requests and Rackspace does not.

This move shows that Rackspace, with its new OpenStack-based cloud services, is bound and determined to compete with the biggest of the big cloud players. In the past the company has seemed reluctant to compete on price alone, instead invoking its “maniacal fanatical support” mantra to justify a premium.”

“Rackspace has always responded to AWS price cuts with statements like ‘but we have great service.’    Looks like the market is demanding the cuts,” said David Linthicum, CTO and founder of consultancy Blue Mountain Labs and a GigaOM PRO analyst.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Migrating media applications to the private cloud: best practices for businesses
• How tomorrow’s mobile-centric data centers will look
• Cloud computing infrastructure: 2012 and beyond

The VMware support is a big deal for OnApp as it helps the company’s service-provider customers better target the enterprise (OnApp already supported Xen and KVM hypervisors, and still intends to support Hyper-V). A new feature called Cloud Boot was introduced to automate the deployment of hypervisors, and there’s a new support console for cloud administrators and end users too.

But it’s the OnApp Storage piece that is particularly critical for the company, Kosten Metreweli, OnApp’s chief commercial officer, told me. This is partly because it solves performance problems for providers, but also because it lays the foundation for OnApp’s upcoming federated compute play.

There are two advantages to this kind of federated storage: it utilizes spare capacity in providers’ data centers, pooling it then slicing up the aggregate into virtual disks, and it also removes the typical bottleneck found in the SAN controller. According to Metreweli, OnApp’s unified approach makes for speedier I/O as well:

“Other distributed storage platforms that have tried to this have required high network bandwidth to work, so we have introduced a clever piece of tech called VM-aware. Because we know where the workloads sit, and we control where the storage sits, we can say at any one point in time we can ensure there’s at least one copy of the data that you’re storing on your virtual disk sitting on the same hypervisor as the compute that’s using it.

“You take away any of the network requirements from a read perspective. You’re getting 95 percent of raw disk performance on what is effectively an enterprise-class SAN, which is pretty unheard-of.”

The content delivery network (CDN) boost is also significant: OnApp’s year-old federated CDN was previously limited to static content and non-real-time “pseudostreaming” — think YouTube – but it now also has a livestreaming capacility.

“We feel that over the last 12 months we’ve validated the concept and proved that this federated CDN capability can work,” Metreweli said.

“We just finished a project with Europe’s largest dance music festival using our distributed CDN to distribute content on a global basis, so now we feel we can take the next step into pushing higher capacity traffic across that CDN. Members of our CDN federation will be able to make more money out of their infrastructure, because there will be more content going over the CDN.”

Others such as VMware and the OpenStack players are also working on the federated cloud idea, and OnApp’s CDN capabilities clearly take on the likes of Akamai, but it’s tricky to identify a direct rival for the sum of what OnApp is doing. Its entrenched network of service provider customers puts it in a good place.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• AWS Storage Gateway jolts cloud-storage ecosystem
• Federated clouds: for when one cloud isn’t good enough
• Infrastructure Q1: Cloud and big data woo enterprises

Ericsson Cloud System, which will hit availability in the first quarter of 2014, targets mobile operators in particular. It incorporates an upgraded Cloud Manager, which is Ericsson’s operations support system, alongside a new Cloud Execution Environment that’s based on OpenStack and the KVM hypervisor.

This all runs on existing Ericsson Blade System (EBS) server clusters and Smart Services Routers (SSR), and the general idea is to enable virtualized environments across the network, from the base station to aggregation nodes and business support systems.

“Cloud services need to be distributed and networks — including computing and storage capabilities — need to be elastic on an end-to-end basis,” an Ericsson statement reads. “This combination will bring a new set of capabilities that doesn’t exist today… It enables distributed cloud capabilities such as computing and storage capabilities in the network, resulting in a better experience when using cloud applications, and more efficient utilization of network resources.

Ericsson’s take on SDN is key to this approach. The company sees most people’s interpretation of the concept as overly focused on the data center, whereas it wants to push the idea of SDN as covering a carrier’s entire network, with operational and business support systems also in the mix.

The benefit, it argues, would be to make resources elastic, to cope with an application’s bandwidth or quality-of-service requirements on-the-fly. By doing so, Ericsson says, the operator can make its own business more efficient, then perhaps use freed-up resources to provide cloud compute and storage services to others.

“There are lots of advantages,” Magnus Furustam, the head of Ericsson’s Core and IMS business, told me today. “Time to market, simplifying operations, but also the innovation that is enabled. By providing a virtualization layer, you make it possible for the operator to [insert] new functionality in the network where it is needed.

“We think it’s important that this is not a disruption –- we’re not asking our operators to throw away their existing infrastructure. Based on that hardware we can upgrade that to support first of all virtualized environments, and if that is not the case we can add hardware, say a board that supports virtualized environments.”

The Ericsson Cloud System will be shown off at Mobile World Congress in Barcelona later this month, where the company will also launch a new “unified” content delivery network (CDN) system called Media Delivery Network.

The new CDN system, designed to help both fixed and mobile operators get into video delivery, combines Ericsson’s existing packet core and radio technologies with new management and service exposure layers. This will let operators do things like select the best CDN in order to optimize traffic, and cache over-the-top (OTT) content — in other words, third-party content — in order to offer the OTT content providers new guaranteed quality-of-service levels.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The Internet of Things: What It Is, Why It Matters
• Cloud and data third-quarter 2012
• Takeaways from the second quarter in cloud and data

Bear in mind that OnApp was only spun out of British hosting provider UK2 a couple of years ago, with software that lets other providers build their own public clouds. The idea there is to help these other hosting providers – OnApp now counts more than 500 of them as customers — ward off the threat that is Amazon, but in the process the company has steadily used that growing federation to diversify into new lines of business.

In 2011, OnApp launched a content delivery network (CDN) based on those service providers’ spare network capacity. There are around 130 points of presence (PoPs) in that network across 40 countries – each provider gets paid for the traffic going over its own PoP, and OnApp gets a 10 percent cut. In 2012, the company took on EMC by doing pretty much the same thing with OnApp Storage, using its customers’ commodity servers to support a distributed storage system that’s controlled by OnApp.

All that is made possible through OnApp’s marketplace and now, flush with fresh funding, OnApp is going to use that marketplace to do the same thing with compute capacity, chief commercial officer Kosten Metreweli told me:

“Adding compute is the most immediate thing. The end customer could now go to [OnApp's customer] and say, ‘I want to spin this up in Tokyo and Moscow’. They can come to our marketplace, buy compute capacity in those locations and also have the application automatically replicated across those locations as well. It makes it much simpler to roll out true global cloud applications.”

There have of course been other marketplaces for compute capacity, such as SpotCloud. On that subject, Metreweli drew a comparison with OnApp CDN competitor XDN, pointing out that OnApp already has a huge customer base brimming with capacity. “The trouble is, they were setting up a market stall in the middle of an empty street,” he suggested.

And he’s not just blowing hot air. In terms of CDN scale, OnApp remains behind market leader Akamai and Limelight but it’s way out in front of Amazon CloudFront and has roughly the same number of PoPs as CDNetworks. OnApp Storage is a newer product, but the company gets to draw on the same customer base there. And those customers can’t hang around these days — not with Amazon breathing down their necks.

“For the majority of service providers, they’d much rather get going with their cloud service, then put their differentiation on top of that,” Metreweli said.

Apart from its compute play, OnApp also intends to use its newfound funding for market expansion – 40 percent of its business is in North America and it really wants to invade non-English-speaking territories. It also intends to turn its storage play, currently bundled with OnApp Cloud, into a standalone product.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Migrating media applications to the private cloud: best practices for businesses
• Infrastructure Q1: Cloud and big data woo enterprises
• AWS Storage Gateway jolts cloud-storage ecosystem

Web site performance is critical because visitors are more likely to leave sites the longer they have to wait for pages to load. For mom-and-pop businesses without any real internal knowledge on how to design fast web pages, this could mean losing business to larger competitors whose sites perform better. And while a CDN won’t cure bad web design, getting a page’s images to load faster can definitely make a difference.

Jason Rosenthal

GoDaddy’s new service, called Website Accelerator, launched on Tuesday for users of the company’s “ultimate” hosting plan and, President of Products and Technology Jason Rosenthal told me, is designed for relatively unsophisticated sites and users. It’s optimized for sites that are fairly straight HTML and maybe a bit image-heavy, rather than dynamic, personalized sites that rely heavily on database interaction. The GoDaddy CDN also starts intelligently caching parts of customers’ web pages with just one push of a button rather than requiring them to go through and tag which content they want housed on a CDN providers servers.

Because GoDaddy handles many domain and web-hosting for many customers, Director of Product Marketing Eric Moyer said, the company is able to optimally route traffic without involving third parties. Early users have seen improvement anywhere between 25 percent and 100 percent, in part because the routing process remains entirely within GoDaddy’s network. Rosenthal said the CDN spans several major population and traffic centers throughout the United States, but that the company will expand it internationally at some point.

GoDaddy is also providing its CDN users with metrics to see how their sites are performing and will offer suggestions on how they might improve performance even more. Moyer said the company partnered with webpagetest.org as the backend for gauging performance improvements.

Really, though, the CDN is just part of a bigger push by GoDaddy to transition from a simple domain and web host into a full-service cloud provider that’s focused on users without software-development or systems-administration knowledge. Yes, it already tried and failed at offering its users something akin to Amazon Web Services and other popular infrastructure-as-a-service offerings, but Rosenthal chalks that up to poor judgment about what GoDaddy’s users really wanted in terms of server-level control.

When it killed its Cloud Servers offering, GoDaddy said it would use those technologies and the lessons learned to build cloud offerings better suited to its users. Rosenthal said Website Accelerator, which has been in development about for about six months, is the product of this plan. “This is really fruit that came out of our Cloud Servers product,” he said. ” … What we heard loud and clear from our customers is ‘I don’t want to know what makes my web site go fast, I just want it to work.’”

Presumably, that means they’re also not interested in looking at other services, such as Yottaa, that claim to be less-expensive CDN options than what giant CDN providers such as Akamai provide. It’s possible they’d work even better, but they would require a greater effort and cost than just sticking with GoDaddy.

GoDaddy actually brought in Rosenthal from Ning.com to help with its transition, and in December it hired former Yahoo and Microsoft executive Blake Irving as CEO. In November, GoDaddy launched a new service to convert customers’ web sites into mobile sites and discussed plans to start offering higher-value cloud applications, as well.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Infrastructure Q1: Cloud and big data woo enterprises
• Quality of the cloud: best practices for ISVs
• Migrating media applications to the private cloud: best practices for businesses

Leighton, who is also chief scientist for the Cambridge, Mass. content delivery network (CDN) provider, will start his new gig January 1, 2013. Last spring, Sagan announced his plans to step down as CEO at the end of 2013. Both Leighton and Sagan will remain on the board and Sagan will stay on as a senior strategy advisor, the company said.

In a statement, Martin Coyne II, the lead director on Akamai’s board said:

 ”When we began evaluating our options for the next CEO of Akamai, our objective was simply to have the best possible leadership team in place for continued growth and innovation in the years ahead, and that’s what we’re doing with Tom’s appointment. We conducted an exhaustive review of the company’s strategy and opportunities as we evaluated a broad range of potential candidates. We are confident that naming Tom is the best and most natural evolution of management responsibility at Akamai.”

The company also promoted two executive vice presidents. Rick McConnell is now president of products and development and Robert Hughes was named president of worldwide operations.  Both will report to Leighton.

Akamai claims a roster of blue chip customers including the BBC, IBM, Microsoft and Hewlett-Packard which also resells Akamai’s CDN as part of its public cloud.

As the nature of web content  has changed, Akamai has had to adapt to handle more dynamic content and has made a series of acquisitions towards that end. Early this year it bought Blaze for its web site optimization expertise just a few months after purchasing CDN rival Contendo. No one expects that M&A  strategy to change.

Akamai competes with Limelight Networks and Level 3 in CDN as well as an array of smaller, more specialized players like Yottaa. Last but not least, public  cloud giant Amazon is both a customer and a competitor to Akamai. Amazon offers its own CloudFront CDN service but utilizes Akamai CDN to speed content delivery around the world.

Just a few weeks ago Akamai and AT&T inked a deal under which AT&T will deploy Akamai CDN servers at the edge of its network and resell Akamai services — in effect bowing out of its own CDN business.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Migrating media applications to the private cloud: best practices for businesses
• 12 tech leaders’ resolutions for 2012
• Infrastructure Q2: Big data and PaaS gain more momentum

By way of background, Backblaze is a cloud storage provider focused solely on backing up lots of data for cheap (like $5 a month for unlimited capacity cheap). In order to do that, it had to build a storage system that could hold massive amounts of data without breaking the bank. As of last July, Backblaze’s architecture had evolved to a point where a 135TB pod cost less than $7,400 to build from scratch.

Understandably, the architecture generated a lot of interest from companies and organizations wanting to leverage it to soothe their own IT budgets, but none of them are Netflix. EMC’s (e emc) Pat Gelsinger said recently that the storage component of Facebook’s Open Compute Project, called Open Vault, isn’t yet ready for primetime because nobody is running — or would run — mission-critical workloads on it. That might be true of Open Vault today — the project just launched earlier this year — but it likely won’t be for long. If you consider a CDN that serves Netflix streaming video mission-critical, the criticism is already invalid for Backblaze’s designs as Netflix has adapted them.

Netflix’s 4U, 100TB server

It’s worth noting, too, that open source hardware isn’t the only piece of the stack threatening legacy storage vendors such as EMC. I’ve heard it suggested recently by someone experienced in building out large-scale cloud infrastructure that the Hadoop Distributed File System has the potential to become the default file system for large infrastructures once it works out some of the limitations around performance and availability. One of the biggest of those limitations — the NameNode– has been eliminated in the latest version of Apache Hadoop and is already integrated into Cloudera’s new CDH4 release.

Can storage deal with the open source disruption?

As with Open Compute’s effects on the server industry, though, open source storage doesn’t need to spell doom for legacy vendors if they’re willing to adapt. One reason is that, at least in the short term, there are still plenty of customers that don’t operate at Facebook or Netflix scale and can afford to pay a premium on smaller deployments that offer the features (and vendor support) those customers demand.

If the server shipments tell us anything, though, it’s that the rise of cloud computing and web giants will ultimately take a toll on the storage market, too. Fewer, but very large, customers will be responsible for a greater percentage of sales, and they won’t necessarily want all the bells and whistles that make enterprise storage products so expensive. And if VMware is correct, even mainstream enterprises will soon want to follow the examples of web giants like Google and Facebook by running relatively dumb hardware managed by really smart software.

If this scenario plays out, storage vendors will have to reassess how they deliver value and earn their money. That might mean adopting open source designs in their own gear while shifting their focus a lot more heavily toward software and services, or perhaps unlocking their storage-management software from the hardware and certifying it to run on open source gear.

Perhaps we’ll get some ideas for what the future storage and markets look like at our Structure conference June 20 and 21, where we’ll dive into the topic with Facebook’s Frank Frankovsy, Netflix’s Adrian Cockroft and VMware Steve Herrod. Whatever the case, it looks like something will have to give.

Feature image courtesy of Shutterstock user Zadorozhnyi Viktor.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Migrating media applications to the private cloud: best practices for businesses
• Infrastructure Q1: Cloud and big data woo enterprises
• A near-term outlook for big data