CloudSigma, which operates out of both Switzerland (Zurich) and the U.S. (Las Vegas), was one of a handful of infrastructure-as-a-service (IaaS) providers that signed up last November for SolidFire’s all-SSD storage system. The result is now here: CloudSigma has ditched all its hard-disk drives and, as a result, it now feels confident enough to offer a service-level agreement (SLA) for performance, as well as uptime.

What’s more, despite the fact that solid-state storage costs about eight times as much as hard-disk, CloudSigma hasn’t changed its pricing – its SSD-based utility service costs $0.14 per GB per month, same as the HDD-based service did. Customers can also pick up the SSD storage service unbundled from CPU and RAM if they so choose.

HPC in the public cloud

According to CloudSigma COO Bernino Lind, the shift to SSD is a major help when it comes to handling high-performance computing (HPC) workloads, such as those of Helix Nebula users CERN, the European Space Agency (ESA) and the European Molecular Biology Laboratory (EMBL):

“They want to go to opex instead of capex, but the problem is there is no-one really who does public infrastructure-as-a-service which works well enough for HPC. There is contention — variable performance on compute power and, even worse, really variable performance on IOPS [Input/Output Operations Per Second]. When you have a lot of I/O operations, then you get all over the spectrum from having a couple of hundred to having 1,000 and it just goes up and down. It means that, once you run a large big data setup, you get iowaits and your entire stack normally just stops and waits.”

Lind pointed out that, while aggregated spinning-disk setups will only allow up to 10,000 IOPS, one SSD will allow 100,000-1.5 million IOPS. That mitigates that particular contention problem. “There should be a law that public IaaS shouldn’t run on magnetic disks,” he said. “The customer buys something that works sometimes and doesn’t work other times – it shouldn’t be possible to sell something that has that as a quality.”

CloudSigma has also resolved another contention point around RAM, Lind claimed:

“A modern CPU can ask for a lot of data because it’s fast and efficient, so it is possible to saturate and make contention on your memory bus. That has been solved with NUMA topology, which is like a multiplexer to get access to memory banks. You get asynchronous access, which means you don’t have contention on accessing the RAM.

“However, public cloud service providers turn this off so the actual instance doesn’t have access to NUMA. We figured out a way to pass on the NUMA topology so, when you run really extensive compute jobs, you won’t hit a kind of contention when you want access to RAM. This is really important for big data workloads.”

In-house stack

Speaking of things that public cloud providers tend to turn off, CloudSigma’s stack – apart from the underlying KVM hypervisor, everything was written in-house – makes it possible to access all the instruction set goodies that are built into modern processors, such as the AES encryption instruction set.

Public clouds may run on a variety of physical hosts that encompass a range of CPU generations, only some of which will have certain instruction sets hard-coded onto the silicon. Providers will often turn off these instruction sets to make their platform homogeneous, but that means losing out on the performance benefits offered by hard-coding. According to Lind, CloudSigma’s stack allows a heterogeneous cloud based on allocation pools – say, one of older Intel chips and another of newer AMD 6380 chips – that customers can choose according to their performance needs.

What does all this mean in practice? Lind cited the example of augmented-reality gaming outfit Ogmento, which recently used CloudSigma’s all-SSD setup to power a mobile, location-based version of a popular title. “They [said] all their I/O-heavy stuff, databases and so on, saw a x8-x12 performance increase,” he noted. “Their entire stack saw a x2-x4 performance increase. That means they need to use less compute power in order to run their system.”

With the budgetary constraints faced by European scientists these days, it’s not hard to see how that same kind of effect could make a real difference in more serious applications too.

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

• Helix Nebula and the future of Europe’s cloud
• Infrastructure Q1: IaaS Comes Down to Earth; Big Data Takes Flight
• Cloud Computing Reaches the Final Frontier

How sure? Well, these are scientists so there’s still a note of caution, but Joe Incandela, a spokesman for one of the LHC experiments, went on-record with a pretty confident statement: “The preliminary results with the full 2012 data set are magnificent and to me it is clear that we are dealing with a Higgs boson.”

However, they’re still not sure what kind of Higgs boson they’re looking at. From today’s statement:

“Having analysed two and a half times more data than was available for the discovery announcement in July, they find that the new particle is looking more and more like a Higgs boson, the particle linked to the mechanism that gives mass to elementary particles. It remains an open question, however, whether this is the Higgs boson of the Standard Model of particle physics, or possibly the lightest of several bosons predicted in some theories that go beyond the Standard Model. Finding the answer to this question will take time.”

It’s not surprising that this task takes time. CERN said a month ago that its storage systems were holding 100 petabytes of data.

The research organization has been working closely with companies such as Yandex to sift through that information in search of unusual events, and in Thursday’s statement CERN pointed out that finding one event means looking through around a trillion proton-proton collisions.

“To characterize all of the decay modes will require much more data from the LHC,” the statement read. For now, the LHC is turned off – it will come back online next year.

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

• Helix Nebula and the future of Europe’s cloud
• Readers weigh in: future prospects for Twitter
• AWS Storage Gateway jolts cloud-storage ecosystem

The new arrangement was announced on Thursday, along with confirmation of Russia’s Yandex becoming an openlab associate in the field of data processing. Huawei’s involvement is a bigger deal than that, as it puts the Chinese firm on a par with Intel, HP, Oracle and Siemens, all of which work particularly closely with CERN to see how their technologies can help with the Large Hadron Collider experiments.

In Huawei’s case, the company is contributing its self-healing UDS cloud storage system for use and validation. UDS is targeting the upcoming exascale (an exabyte is roughly a million terabytes) era with a mass object-based storage infrastructure that uses ARM’s energy-efficient processor architecture alongside cheap SATA disks. It also offers Amazon S3 API compatibility and claims eleven-nines (99.999999999 percent) reliability, so users theoretically don’t need to back up data stored in a UDS-toting cloud.

UDS provides a bit of insight into how openlab works. Huawei first delivered a 384-node version of UDS to CERN in early 2012, after which the researchers played around with it for three months. In September of that year, Huawei released UDS to the general enterprise market (in more normal eight-node configurations). The benefits for both sides of this partnership are clear: CERN has to push technological limits in order to handle the very big data generated by the LHC, and Huawei gets both valuable feedback from the researchers and a glowing report card to show off to the wider world.

As for the next steps in this partnership, CERN has now hired two computer scientists to work with Huawei on its implementation there, and more UDS storage systems will be deployed at the Swiss facility in the next few months.

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

• CES 2012: a recap and analysis
• 9 Companies that Pushed the Infrastructure Discussion in 2010
• Infrastructure Winners and Losers of 2009

The technology in question is called MatrixNet. It’s a machine learning tool that helps Yandex rank search results for relevancy, but in this case CERN is testing it out alongside other proprietary Yandex data-processing technologies in order to pick rare but important events out of the masses of data produced in CERN’s experiments. Essentially, it is being used to more accurately establish statistical relevance, so that CERN can call something a ‘major discovery’ with more certainty.

The research facility is already using MatrixNet to sift through data on B-meson decay analysis (this is part of the quest for supersymmetry, which tests the boundaries of the Standard Model of physics), but now CERN will also see whether it can help across other aspects of its work.

To accomplish that, Yandex has been brought in as an associate member of CERN openlab, the mechanism CERN uses to collaborate with private-sector IT firms such as Oracle and HP. This is actually an expansion of Yandex’s work with the research facility, as it has already been supplying it with computing resources since 2011. The company also came up with a specialized search tool last year for the data emanating from one of the Large Hadron Collider experiments.

CERN openlab head Bob Jones said in a statement:

“Today’s physics deals with large datasets that need to be properly processed and interpreted. New discoveries are all but impossible without meticulous data analysis. To this respect we are very interested in using and testing the services and technologies Yandex is developing.”

This is sort of a sponsorship deal, giving Yandex added respectability, but it’s also win-win from a technological standpoint — after all, we’re talking about machine learning here. As Yandex’s Andrey Ustyuzhanin put it: “Being a CERN openlab associate member opens up new opportunities for contributing to the organization’s projects and gives us a wide access to the LHC experiment data. These datasets allow us both to improve our technologies and contribute to advances in physics.”

The Russian web giant previously had to supply CERN with specialists in order to work with MatrixNet, but it said in Monday’s statement that it is developing “a user-friendly service that will give physicists an opportunity to use Yandex’s machine learning technology on their own”. It’s not clear whether this is to do with the fact that MatrixNet is proprietary, but it is worth pointing out that Yandex’s willingness to even let outsiders near its secret search sauce sets it apart from chief rival Google.

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

• Takeaways from the second quarter in cloud and data
• The importance of putting the U and I in visualization
• A near-term outlook for big data

Speaking at GigaOM’s Structure Europe conference in Amsterdam on Wednesday, Robert Jenkins of CloudSigma explained that “when you give people the data, you don’t know the end result but it comes up with a lot of positive outcomes.”

Jenkins gave the example of the World Health Organization tapping into data from the European Space Agency in order to improve mosquito eradication efforts in Nigeria. These type of collaborations are possible in part thanks to CloudSigma and other private companies that are helping to supply the computing power required to collect and parse data across Europe.

Ensuring that publicly funded-research produces public data has also resulted in an eco-system of small and media enterprises, according to Bob Jones of CERN who joined Jenkins at the event. Jones said this eco-system and the power of the cloud is producing the type of inter-disciplinary algorithms that are necessary to solve current science and policy problems.

Moderator Kris Tuttle, who is CEO of SoundView Technology Group, asked how to determine a common resource unit or module when working in such a vast “federated cloud.” Jenkins said the answer is to focus on deliverables rather than resource units, offering the example of a gene sequencing project in which the customer simply cares about how long it will take to get the job done rather than the specifics of the implementation process.

The speakers also noted that working in the European science cloud also provides practice and credibility for companies that want to serve private sector industries like finance, media and biotech.

Finally, the discussion produced some mind-boggling science stats like the fact that the Helix Nebula is now using 200,000 CPUs to continuously churn data. Jones also noted that maintenance windows are 18 months due to the fact that much of the work is occurring near a temperature of 270 kelvin — meaning that machines have to be slowly warmed up from a state of near absolute zero.

Check out the rest of our Structure Europe 2012 coverage here. A video recording of the session follows below.

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

• Helix Nebula and the future of Europe’s cloud
• A near-term outlook for big data
• Infrastructure Q4: Big data gets bigger and SaaS startups shine

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

• A near-term outlook for big data
• Infrastructure Q1: IaaS Comes Down to Earth; Big Data Takes Flight
• Cloud computing 2013: how to navigate without a map

We’ll discuss many of the finer points of European cloud computing and web infrastructure at our Structure: Europe event Oct. 16 and 17 in Amsterdam, both business and technological. To whet your appetite, though, here are seven reasons why Europe is a lot more important than many people might think.

1. Clean, innovative energy

Northern Europe is turning out to be a testing ground for some of the most-innovative data center designs around thanks to its cool temperatures and abundant supply of renewable energy. Most notably, there’s Google’s seawater-cooled facility in Finland, and Facebook’s hydropowered plant in Sweden. Iceland is popular too, hosting Verne Global’s 100-percent renewable converted NATO hangar, and home to GreenQloud — a cloud provider running wholly within Verne Global’s facility and another all-clean data center by Thor Datacenter.

2. Eastern Europe’s talent pool

Eastern Europe has become an important geography for the IT world in general, boasting a glut of programmers, software engineers and other workers with skills that today’s companies really need. Not only is the region a hotbed for outsourcing, but it’s also the home to a growing number entrepreneurs doing innovative cloud and big data startups. Among them is Nginx, the popular web server created and since commercialized by Moscow resident Igor Sysoev .

3. London’s financial center

There are probably many reasons Dublin, Ireland, is emerging as a data center hotspot — including acting as the European homebase for Amazon Web Services’ and Microsoft’s cloud businesses — and one of them is almost certainly to be as close as possible to London without having to deal with that city’s severe power shortage. But London is an important market, especially when it comes to the financial world, an industry that spends billions on IT each year and that cloud computing providers want desperately to tap into.

4. CERN

One really needn’t say more than “Tim Berners-Lee” when describing CERN’s influence on cloud computing, because the cloud as we know it wouldn’t be possible without the web. But CERN also created a little thing called the Large Hadron Collider, which aside from discovering the Higgs boson, is the impetus of possibly the world’s most-impressive research network — spanning dozens of countries and supercomputers worldwide and generating around a petabyte of data per second. CERN is a big supporter of open source cloud efforts, including OpenStack, OpenNebula and the Helix Nebula research cloud in Europe.

5. OpenNebula

Led by Spanish computer scientist Ignacio Llorente, OpenNebula is a fairly popular open source cloud platform that rivals the work being done, largely in the United States, by the OpenStack and CloudStack projects. The project has been around since 2005, and claims a handful of large companies and European research institutions as users. Although it doesn’t have the big-name backers of the other two, it should remain viable for a long time because of its rather large user base.

6. One-third of Twitter’s firehose

DataSift, which is headquartered in Reading, England, is one of three companies (along with Gnip and Topsy) certified to resell the all the billions of data points streaming from Twitter every day. Social media, and Twitter especially, are a huge focus of corporate analytics efforts, and anyone that can capture and analyze all the world’s tweets is a kind of a big deal.

7. Individual rights

American cloud providers might not always agree with the European Union and its individual countries when it comes to data governance laws, but every country’s citizens should pay European governments some respect. Their regulations on data privacy — such as the prohibition on Facebook’s facial-recognition feature in Germany — help keep privacy in the forefront of companies’ minds as we release more and more personal info to web-based services.

Privacy image courtesy of Flickr user striatic.

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
• 9 Companies that Pushed the Infrastructure Discussion in 2010
• How to Make Cloud Computing Greener

Large Hadron Collider

Scientists have a limitless hunger for computing power and storage. That’s why three European agencies — CERN, the force behind the Large Hadron Collider; the European Molecular Biology Laboratory; and the European Space Agency — are cooking up a European science cloud to handle their compute-intensive workloads.

The Helix Nebula Science Cloud aims to bring enough firepower to solve very hard problems and deal with tons of data churned out by the second. As an example of the size of the task, CERN alone stores about 15 petabytes of data per year, uses 150,000 CPUs continuously, and writes data at 6 GB per second.

“Physicists can consume as much compute power and storage as we can give them. There is so much data and so many ways to explore and mine that data, that we’re always looking for new techniques to increase that total capacity,” said Bob Jones, the head of CERN’s OpenLab in an interview. CERN, or the European Lab for Particle Physics, is based in Geneva and funded by European Union nations.

The project is backed by commercial partners including Atos, Capgemini, CloudSigma, Logica, Orange Business Systems, SAP, Telefonica, the Cloud Security Alliance and the OpenNebula project.

Up until now CERN has been using publicly owned infrastructure — 150 data centers in all — that contribute resources for LHD data. “We want to see if we can use commercial infrastructure as an additional resource,” Jones said. Like any organization, the group is trying to build in flexibility and maximize price performance.

Discussion around the project started last summer, and the group decided to use the new infrastructure initially for three projects. CERN will use the infrastructure to handle the Atlas Experiment, a particle physics project, at LHC. The European Molecular Biology Lab will use it in a new project to simplify the analysis of large genomes and the European Space Agency, working with two partners, will build an application to study earthquakes and volcanoes.

The first stage of the project is a two-year pilot building a cloud to handle those three applications.

This is a Europe-only cloud. “The initial drive is from these European research labs,” Jones said, “and one reason it’s ended up this way is this whole question of data security and access. We have to be certain wherever we start to use these resources that we have appropriate legislative backing.”

There has been a movement to build Europe-only clouds, with proponents usually invoking the U.S. Patriot Act as rationale. Because some European nations –notably Germany — have tougher privacy laws than those in the U.S., local companies want to make sure data in their data centers cannot be turned over to U.S. authorities.

As the prototyping continues, Jones said the group will look into bringing other research organizations and cloud providers aboard.

Photo courtesy of  Flickr user Image Editor

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

• Helix Nebula and the future of Europe’s cloud
• A near-term outlook for big data
• Will cloud computing push the BRIC market to the front?

CERN Scientists monitoring the LHC

In the age-old quest for humankind to discover the secrets of the universe, humankind has progressed today as the Large Hadron Collider successfully smashed protons by zipping the subatomic particles around a 17-mile loop at speeds that were about 99 percent of the speed of light. The goal, nothing less than to figure out how the universe works and what holds it together. What does this have to do with broadband, cloud computing or wireless? Frankly, I have no idea, but I’m convinced that such efforts advance technology, and act to inspire future generations of engineers and scientists.

Indeed, while it may not be apparent from my normal coverage, the LHC built by CERN represents why I spend my days writing about technology — not because I’m excited to play with the latest gadgets, but because I value the spirit of curiosity and discovery that leads scientists to spend $16 billion to build something that may (not will, but may) give us an inkling about how the universe works. And if it enables us to teleport, that’s cool, too. So if you’re interested (and how can you not be?), step away from the iPad coverage and check out these links:

• Official CERN Announcement of Success
• Time Traveling Particles Out to Destroy the LHC
• The LHC Will Kill Us All
• A Photo Tour of the LHC
• CERN’s Twitter Page and its live webcast
• Comic from Jorge Cham explaining LHC
• Because I love it: The Large Hadron Rap
• The BBC’s guide to the LHC
• TED video on LHC (embedded below)

http://video.ted.com/assets/player/swf/EmbedPlayer.swf