The Do-it-Yourself Datacenter

By By Dennis Barker, GRIDtoday

May 5, 2008

The idea of financing your own private datacenter off your credit card isn’t so crazy anymore. Renting CPUs is nothing new: At least two gigantic companies with very famous Web sites will let you run your applications on their metropolis-sized server farms, with fees within reach of many businesses.

But the model they offer doesn’t work for everyone.

What if you want a datacenter built specifically to maximize your application, using the software and hardware components you want, and costing you only when you use it? Can you put hat tailor-made, on-demand datacenter on your credit card?

Si, usted puede. (Yes, you can.)

Several providers are taking advantage of technology developed by 3Tera to offer customers whatever level of grid computing power they need, on a pay-as-they-go basis, but allowing a level of flexibility you don’t get with the more traditional or gigantic hosts. 3Tera set out in 2004 to enable companies to run scalable Web applications on grids of commodity servers, on a utility basis. In 2006, 3tera went live with AppLogic, its grid operating system designed specifically for scalable Web applications.

“It’s not a grid in the traditional terms,” says Bert Armijo, senior vice president of sales, marketing and product management. “It’s a grid in terms of organization of resources, the ability to rack and stack at will and spread apps across hardware, and it separates the app from the hardware, so it’s an excellent tool for utility computing. But it’s also a meta OS, it runs other OSes within it.”

AppLogic mimics traditional hardware facilities, so you can run your existing apps on top of it, Armijo says. One of the biggest user benefits of the 3Tera architecture is that businesses can use the operating systems, middleware, applications, open source tools, security methods, and vendors they choose, and not what the host prefers, he says.

In 3Tera’s system, you’re running virtual appliances. “Each one has everything it needs to run on the grid, like OS code, application code, database schema, file storage, load balancer, whatever makes the app what it is, plus all the configuration information,” Armijo says. “This is why apps can scale from a tiny slice of a server to hundreds of servers on demand, be replicated on demand, or deployed in different locations without the customer having to make any modifications.”

Because you can send these app packages anywhere with one command, it makes it easy and convenient to move an application between networks or locations — a pretty handy backup and redundancy scheme. Depending on your host provider, you could have copies running on datacenters in other parts of the country.

Armijo demonstrated over the Web the visual tools used to define an application in AppLogic, dragging and dropping elements such as Web servers and firewalls from a catalog. “Our infrastructure editor uses a visual metaphor, so configuration is simple, and you don’t have to do things like login and edit tables or worry about syntax.” A visual interface also is used to set the amount of resources each process gets, and there’s a dashboard for monitoring the grid and managing applications.

“We’re giving people the ability to define their application exactly as they want it and move it onto the grid — on resources they don’t need to own,” Armijo says.

Your Datacenter, Your Way

ENKI, a cloud computing provider, consultancy, and owner of many servers, adopted 3Tera’s grid OS as the foundation for its meter-based virtual datacenters. (ENKI isn’t just a customer; it’s also a 3Tera partner.) The company, which started out as a shop of virtualization experts, moved into managed hosting, then met up with 3Tera. “We had a similar vision of where computing was going,” says Eric Novikoff, ENKI’s chief financial officer, “and their technology would let us cover a large percentage of the areas we wanted to serve, including deployment, scalability, high availability, storage, security and so on, and offer it on a utility-billed basis.”

“People come to us with a mental image of a physical datacenter but want to know if they can deploy it in our cloud offering. In most cases you’d have to do extensive work to convert to the cloud,” Novikoff says. “But inside AppLogic, we just draw it [dragging and dropping components and connections], press a button, and the virtual datacenter looks like the physical datacenter. This is important to many developers because they’ve written software using tools like Ruby on Rails or Java that have certain assumptions about how things will work. That should be determined by the application writer, not the service provider.”

ENKI makes AppLogic available in one of three ways: (1) a customer with its own stack of enterprise servers can lease the software and have its own staff, or ENKI’s staff, manage the virtual datacenter; (2) a customer can lease servers from a 3Tera managed services provider but serve as its own datacenter manager (the be-your-own-cloud-provider option); or (3) ENKI can be the cloud service provider, on a pay-for-use basis. ENKI will cloud-consult no matter where the customer hosts.

There is a guy financing his startup on his credit card, and he’s running his application on his own private virtual datacenter in ENKI’s cloud. “He got a network of Linux servers from us and installed his software on each one, as though they were co-located physical servers in a datacenter, and he administers them as if they’re physical servers,” Novikoff said. “But he’s really only using a quarter or an eighth of a physical server, and that saves him money. He’s paying only for what he uses. If his load goes up, we can scale up more servers.”

Another customer operates a Web site, ubuket.com, that’s very heavy in video and audio files, which all are stored in ENKI’s datacenter. “Because they’re using increasingly more storage and more CPUs, they keep changing their architecture to improve performance. With AppLogic’s tools these changes are easy to make,” Novikoff says. “They can also test changes by deploying the software in the grid, then when it’s ready push a button to redirect it to the live site.”

Add a Layer of Grid

Layered Technologies is a hosting provider and computing utility that owns about 12,000 servers in eight datacenters. “We combine our hardware and our management of the service with the AppLogic interface and offer clients hosting on our different grids a variety of server sizes,” says John Pozadzides, chief marketing officer. “We call these virtual private servers.”

There are two things customers like about these grid-based VPSes, he says. First: redundancy. All the data written to the individual machine also is mirrored on other hard drives within the grid. Once a VPS is running, the grid monitors it and migrates to another machine if there is a node failure or other stoppage. Second: adaptability. “A small company that doesn’t really know how to anticipate its growth can buy a VPS in the grid and if its requirements increase, we turn up the juice,” Pozadzides says. “You need more RAM, you make a phone call, we turn up the RAM. In other hosting situations, you’d have no option but to add another box.”

Layered Tech sells its grid in one of two ways. Users can buy individual servers and it will use AppLogic to provision them. “Or we will turn over the AppLogic interface to you along with a certain number of servers. We call this a virtual private datacenter. It’s an enterprise-grade hosting solution. You might have four physical nodes, each with quad-core processors, and 8 gigs of RAM. We provide AppLogic and now you have a grid with 16 processor cores. If you need more machines, you add them to the grid. You could have a hundred servers in there and you’d manage them all through one simple graphical interface.”

Layered Tech’s virtual private datacenters are enterprise-grade, Pozadzides says, but managing them is easier than traditional systems. “There’s no hardware to configure or maintain,” he says. “When you run an application, it configures all the infrastructure pieces on demand and runs your code on top. When you stop the app, AppLogic knows to turn off or reassign any resources no longer needed.”

It’s so Easy

By melding 3Tera’s technology with their hardware and hosting expertise, providers like ENKI and Layered Technologies can offer pay-per-use grid capabilities to businesses that could never afford to build their own datacenter. “We’re trying to make it easy for companies to move their Web applications to the grid,” says Armijo. “AppLogic is designed for their types of intensive workloads.”

ENKI’s Novikoff agrees. “What I see all of us trying to do,” he says “is [lower] the barrier to having a monetizable Web application.”

Sounds priceless.

Get Off of My Cloud

At the recent Web 2.0 Expo, 3Tera decided to open things up with its Cloudware architecture and initiative to promote “open cloud computing.” (3Tera’s marketing phrase is “Cloud Computing Without Compromise.”) The architecture will cover all the building blocks of a cloud (e.g., computing, storage, security, etc.) and make them work together. Cloudware incorporates AppLogic, but “we are decomposing AppLogic into its primary parts, so other folks can add their unique ideas and capabilities to the cloud,” Armijo says.

3Tera will publish specs that developers need to “publish” their resources. The idea is that developers will make their technologies available to all, stored in a global catalog. A company like F5 Networks, for instance, could make its load-balancing technology accessible in the open cloud, Armijo says. This could be a real break for small businesses that don’t have a lot of money to spend. It could also turn the cloud into a hothouse of innovation, like that old thing called the Internet.

“The problem with cloud computing,” Armijo says, “is that most vendors, when they say ‘cloud,’ mean ‘my cloud.’”

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