Cloudnumbers Expedites Cloud-Based Supercomputing

By Tiffany Trader

February 14, 2012

Configuring a cluster on a public cloud infrastructure such as Amazon Web Services potentially requires a lot of work. The various steps involve setting up the machines, dealing with the security keys, installing the applications, negotiating the administration, and more. Most HPC users would prefer to avoid this time-consuming process if possible. That’s where German startup Cloudnumbers comes in. The company provides everything the customer needs to get their HPC applications up and running in a matter of minutes. I recently had the chance to speak with Markus Schmidberger, Senior Community Manager for Cloudnumbers.com GmbH, to find out how.

The history of the Cloudnumbers goes back to Munich circa 2010, when company founders Erik Muttersbach, Markus Fensterer and Moritz v. Petersdorff-Campen began developing their idea for an HPC cloud environment. They contacted Markus Schmidberger in October 2010 after seeing his research on the topic, and further discussed the feasibility of moving HPC workloads to the cloud.

By the end of 2010, the group had written up a business plan and begun implementing a prototype. In March 2011, Cloudnumbers was founded with investment from angel investors. At that time, the company relocated from Munich to Berlin, Germany’s capital city. After an initial testing period of about two months, the company officially launched on July 4, 2011, providing a Web interface for computer clusters in the cloud. As of now, the company has over 1,300 registered users.

If you go to their website, you may notice a beta tag on their offering, but Schmidberger assures me this is a fully-functioning stable release. The beta designation has more to do with German law than any product short-comings. Based on German law, they would have to provide a level of security that is impractical at this stage. The beta label just makes it easier for them to operate in Europe, Schmidberger explained.

In case you’re concerned about security, the core functionality is the responsibility of the IaaS provider, but Cloudnumbers has added an extra security level on top of what Amazon uses. This means that all data are encrypted with 256 bit encryption and all connections between the local hardware and between the machines use SSL encryption. Despite these high levels of security, performance has not been affected in any significant way. It bears mentioning that all aspects of user authentication, personal data, passwords, billing and credit card information are located on a secure server in Germany and not in the cloud, in full compliance with safe harbor requirements.

Cloudnumbers does all the configuration of the computer cluster for their customers. Prospective users can point their browsers to the Web interface, where a five-step process results the configuration of a computer cluster. At that point, the cloud provider starts up the cluster and performs the complete install with all required applications. After that, the customer can access the cluster through any Web interface. “Everything is pre-installed so they don’t need any administration. Basically everything works out of the box,” Schmidberger explains. Setup takes about five minutes, and the entire configuration process takes between 6-10 minutes.

To use the service, customers only need to register with Cloudnumbers. They do not need a separate account with the public cloud provider, so there is only one bill. Every registered user gets five hours free, and after that the service costs about one dollar per hour per CPU. This is slightly more than what Amazon charges, but Cloudnumbers is providing something extra, a preinstalled cluster with preinstalled applications. While Cloudnumbers uses Amazon Web Services as their main Infrastructure-as-a-Service platform, they are currently adding resources from RackSpace and Eucalyptus, and are actively seeking to partner with smaller cloud providers as well.

Cloudnumbers offers pre-installed computer clusters in the cloud with pre-installed applications. Depending on their needs, users can select from R, OpenFoam, Python, Fortan, C/C++, as well as Gromacs, BLAST, Freemat and Perl. These come with the associated libraries for parallel computing, such as all the MPI libraries and so on. “We do all the hard work for setting up the computer cluster in the cloud. It’s integrated in our interface,” notes Schmidberger.

In analyzing the behavior of their customers, Schmidberger finds that about one-third require one shared-memory machine with 8-cores and 64 GB of main memory, and the other two-thirds set up computer clusters with about 10-15 machines. If you look for competitors, you may come across Cycle Computing, which offers very large clusters. In Schmidberger’s experience, Cloudnumbers’ customers haven’t required machines this large. Their customers tend to be of the small-to-medium sized variety, and are thus well-suited to small-to-medium sized clusters. They care about getting their high-performance applications to run on these clusters, Schmidberger’s notes, but are not as interested in the learning the administration side.

With that in mind, it’s perhaps not surprising that many of their customers require consulting services to help ease the transition to the cloud. They want to know how to get their code running in the cloud and then how to get the best performance. Initially, Cloudnumbers only planned to offer the cloud product, consulting was not part of the original business plan. But after receiving numerous requests for assistance, a consulting service was added in October of last year. The bulk of the service involves simple programming support in the way of code optimization, tips and tricks. In most cases, the actual coding is performed by the customers.

“We provide the complete HPC-as-a-Service package for our customers,” Schmidberger says. “We have the support, we have the consulting, and we have the interface and the resources, which they can use.”

To illustrate the way the consulting process works, Schmidberger cites one of their most well-known customers, Gazprom Germania, a holding company for a range of international oil and gas companies. Gazprom has a lot of financial code for risk analysis, based on the statistics software R. Cloudnumbers worked with Gazprom to parallelize the code, with the actual implementation being performed by Gazprom statisticians. After that, Gazprom simply logged into the Cloudnumbers test account, confirmed everything was in order, then switched to on-demand billing and started working on the Cloudnumbers machines.

While some customers require the complete support package, most just log on and get a bill every month. For these, Cloudnumbers doesn’t even know what workloads they are running, nor do they need to know.

The initial premise for the company was to close the gap for high-performance computing in the cloud in Europe. To get into the market in Europe, and to that point, their majority of their customers are from the EU, but they have several from the US, several from China and India. The numbers break down to roughly one-third from Europe and the rest from everywhere else. The diverse customer base includes academia, industry, the finance sector, biology firms, and machine engineering outfits.

Asked whether customers have had any difficulty transferring initial data sets to the cloud, Schmidberger responded that bandwidth was not a problem since, again, the solution is designed for small-to-medium sized companies with average data sets in GB range, not on the level of TB. In fact, Schmidberger estimates they’ve never had a customer with more than 1-2 TB of data. So far, not one of their customers has complained that they’ve had a problem loading data from their local machine into the cloud.

On the somewhat sticky topic of whether HPC workloads are suitable for the cloud, Schmidberger is convinced of the merits of this approach. “While we will continue to require supercomputers for huge analysis runs,” he says “for the small-to-medium sized analysis, HPC works very well in the cloud.”

To get the customer perspective on this new cloud service, I reached out to Dr. Karsten Knothe, IT Project Manager for Gazprom Germania, mentioned earlier. Before moving to Cloudnumbers, Gazprom ran their workloads in-house on a simple desktop-PC with multicore processors. When they started thinking about transitioning to the cloud they briefly considered going directly to Amazon, but they needed more functionality. “For us it’s important to get not only a naked virtual machine but a configured cluster with the needed software installed,” Dr. Knothe comments.

Prior to the cloud transition, one calculation run would take more than 12 hours. “Now we do one calculation in about 15 minutes,” notes Dr. Knothe. Not only do they save time, but the solution allows them to do additional analysis by changing the parameters of their calculation.

Despite being satisfied with these benefits, Dr. Knothe mentioned that he would like to see a few additions to the product. Specifically, he has requested additional communication functions, so that starting and stopping sessions and finishing calculations would trigger an email notification.

Looking ahead, Schmidberger was excited to announce that Cloudnumbers is currently developing a hybrid cloud solution, which offers the promise of “perfect scalability.” Due to privacy or security concerns, many companies have chosen private clouds to run their analyses, but very often those companies have some workloads which would benefit from the cloud. A hybrid approach allows these companies to extend their private cloud with public cloud resources. Currently, the company has a prototype connection in place with Eucalyptus and Amazon, with a connection to OpenNebula in the works as well. A final release should be available soon.

“We set up a very generous framework in the background, which gives us the opportunity to connect to many different Infrastructure-as-a-Service providers,” Schmidberger stated. “So if we get a request from any customer that they have a private cloud running on whatever, we will be able to connect to this cloud within two months.”

As a new company, Cloudnumbers is proactive in seeking customer feedback, which should be good news to Gazprom.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire