Datacenter Expansion in a Can

By By Dennis Barker, GRIDtoday

July 21, 2008

The treasury of truck-drivin’ lyrics has yet to include any reference to hauling a datacenter on the back of a rig. But the way things are going, it won’t be long before some updated Red Sovine sings about carrying crates of gigahertz down Interstate 95.

HP is now offering customers the chance to have their datacenter built inside a 40-foot shipping container, instead of a more traditional stationary structure. Not a novel idea — Sun rolled out its Black Box in 2006 — but one with some obvious benefits. For starters, there’s no need to construct a new building or apply an extreme makeover to your current space. Bring in the brick and mortar, and you’re talking at least a year before your new digs is ready. HP says it can deliver a containerized datacenter within six weeks.  

HP’s new Performance Optimized Data Center, or HP POD, is essentially 4,000 square feet of IT real estate filled with custom-configured equipment of the customer’s choice and housed inside one of those big steel boxes often seen on the backs of 18-wheelers and in mobster movies.  

The HP container can support as many as 3,500 compute nodes, or if you wanted it more as a place to store 12 petabytes of your stuff, 12,000 hard drives. There’s room for 22 standard 19-inch 50u racks. (That’s the 40-foot version; a 20-foot version is also available.) To get the maximum compute density that HP claims, you’d need to fill those racks with HP’s newish ProLiant BL2x220c G5, which combines two Xeon-based blade servers into one blade. The 220c doubles CPU capacity while consuming about 60 percent less power than a comparable blade, HP says. But you don’t have to fill the racks with HP equipment.

“We’ll support Dell, IBM, Sun, Cisco, whatever fits in a standard rack,” says Steve Cumings, director of infrastructure for HP’s Scalable Computing and Infrastructure Organization. “You can use all the normal equipment you’d expect in a datacenter. Customers told us, ‘Don’t have your container limit the kind of IT we can put in place.’ People don’t want to be limited to a certain type of equipment or certain brand.” Cumings says HP lets companies mimic the environment they already have in their datacenter. “That way managers don’t have to train people to work in a POD. IT can manage the POD with the same software, the same tools. The POD will plug right into your usual management software. It’s really an extension of your current datacenter.”

HP assembles, configures and integrates third-party equipment if you like, and tests each POD through its Factory Express service. Inside the container, racks are “anchored top and bottom and shock-absorbed for shipping,” Cumings says. “Inside the POD you have your cold aisle and hot aisle, and full access just like you’d have in your regular datacenter. We can deliver everything in one POD, including all the power modules and UPS battery and transformers, or you can have all your utility modules in a separate, 20-foot container.” HP says the separate-box approach is better for two main reasons: the power gear is more temperature-sensitive and should be in a cooler unit, and people who service power hardware might not have the same security clearance as the people who service the IT hardware.

After trucking, installation and deployment, HP will handle whatever the customer needs, including day-to-day management tasks. “There’s a portfolio of services that goes around all this,” Cumings says. “Our design and data center strategy group has a ton of expertise. They’ll sit down with you and help develop a plan. They’ll do site assessments, arrange for site prep, take care of all the prerequisites, and even manage it for you. We’re not just dropping off a container.”

Although the big metal box eliminates the time and cost consumption of construction, HP does not see PODs replacing traditional data shacks. “The POD is a complement,” Cumings says. “Customers are still going to build out some brick-and-mortar space. The POD is an economic and efficient way to support a really intense IT environment or to get more capacity really quickly.”

“We still hear from customers that power, cooling, and space limitations are still problems in their datacenters. Out of power capacity, out of cooling capacity, out of space,” Cumings says. “The thing that exacerbates that is there’s a lead time in building out additional space. It can take up to 18 months to build and fill a typical datacenter, and then there’s a lag time in reaching the capacity limit. People see the need, and then it takes time to get the capacity they need. The POD goes a long way toward solving that problem because you’ll have your expansion within six weeks, delivered, tested, and ready to turn on.”

According to the EPA, energy costs for datacenters are doubling every five years, and with approximately 50 percent of a center’s operational costs related to power, cutting that electric bill is a business necessity. “Customers have been telling us in the past year that energy costs are having more impact than the cost of servers and other IT hardware,” Cumings says. “We can design the POD for increased energy efficiency. You’re going to gain some of that just by going with the BL 2x220c [blade server] because it draws less power than competing servers.” (HP says the 2x220c consumes about 60 percent less energy than a comparable machine from Dell.)”

There’s another way the truckable datacenter could cut utility bills, Cumings says: “It can go wherever the customer wants it to go, so there’s the option of deploying the POD in an area where they can get lower energy rates. If you’re located in a place where utilities are really high, you’re not locked in to that location. You can think about setting up in a part of the country where electricity is less expensive.”

The datacenter-in-a-crate would work for a wide range of organizations, and HP says it has seen interest from financial, healthcare, manufacturing, and collocation companies. Many prospects have more than one datacenter already. But they all have one big thing in common: they’re exceeding capacity and need to expand quickly.

HP would not talk POD pricing, but Cumings did say “there’s not a price premium for the POD itself. We expect it will typically have a lower total cost of ownership than a brick-and-mortar equivalent. It’s going to be more energy-efficient, and certainly more space-efficient.” PODs will be available in the United States in October, and elsewhere next March, HP says.

View from a Real Datacenter

We asked a man who designs and manages datacenters for a living what he thinks of HP’s POD. John Engates is chief technical officer at Rackspace, a large IT hosting services provider with eight datacenters around the world. (Rackspace uses a mix of servers, including HP, Dell and its own house blend.)

Rackspace would consider adding a POD “because we’re interested in any solution that would buy us more efficiency and would be more cost-effective in the short or long term,” Engates says. “We are looking for the sweet spot of power efficiency, space efficiency, acquisition cost, and cost of ownership. As we move into the era of cloud computing, solutions like these become more and more appealing. When you add servers a few or a dozen at a time like a traditional enterprise does, it’s hard to utilize a system like this. But when you start buying servers hundreds or even thousands at a time, it becomes more likely that these containers would make sense. In the cloud, you could essentially treat the entire container as a black box and likely not even service failed components inside until you crossed some threshold, maybe 10 percent failed components or something like that.”

“The coolest — no pun intended — part about these containers is their ability to truly isolate the hot and cold aisles of the datacenter and channel the cool air exactly where it needs to go,” Engates says. “That can be difficult in a traditional datacenter. These containers essentially turn almost any datacenter, or even a warehouse, into a well-designed environment for efficient cooling of server equipment. Anything we can do to cut down on the energy consumed by the equipment, the heat produced, or the amount of energy required to cool the data center is worth pursuing. The containers primarily help better manage the cooling aspects, which is certainly a significant cost.”

HP’s POD is similar to IBM’s Portable Modular Data Center in that both offer a 40-foot container and support third-party hardware. But HP, with its dense hardware, currently has the lead in how much IT can be packed into a pod — and also how quickly it can do it: HP promises to have customers up and running within six weeks; IBM promises 12 weeks.

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!

2024 Winter Classic: Meet Team Morehouse

April 17, 2024

Morehouse College? The university is well-known for their long list of illustrious graduates, the rigor of their academics, and the quality of the instruction. They were one of the first schools to sign up for the Winter Read more…

MLCommons Launches New AI Safety Benchmark Initiative

April 16, 2024

MLCommons, organizer of the popular MLPerf benchmarking exercises (training and inference), is starting a new effort to benchmark AI Safety, one of the most pressing needs and hurdles to widespread AI adoption. The sudde Read more…

Quantinuum Reports 99.9% 2-Qubit Gate Fidelity, Caps Eventful 2 Months

April 16, 2024

March and April have been good months for Quantinuum, which today released a blog announcing the ion trap quantum computer specialist has achieved a 99.9% (three nines) two-qubit gate fidelity on its H1 system. The lates Read more…

Mystery Solved: Intel’s Former HPC Chief Now Running Software Engineering Group 

April 15, 2024

Last year, Jeff McVeigh, Intel's readily available leader of the high-performance computing group, suddenly went silent, with no interviews granted or appearances at press conferences.  It led to questions -- what's Read more…

Exciting Updates From Stanford HAI’s Seventh Annual AI Index Report

April 15, 2024

As the AI revolution marches on, it is vital to continually reassess how this technology is reshaping our world. To that end, researchers at Stanford’s Institute for Human-Centered AI (HAI) put out a yearly report to t Read more…

Crossing the Quantum Threshold: The Path to 10,000 Qubits

April 15, 2024

Editor’s Note: Why do qubit count and quality matter? What’s the difference between physical qubits and logical qubits? Quantum computer vendors toss these terms and numbers around as indicators of the strengths of t Read more…

MLCommons Launches New AI Safety Benchmark Initiative

April 16, 2024

MLCommons, organizer of the popular MLPerf benchmarking exercises (training and inference), is starting a new effort to benchmark AI Safety, one of the most pre Read more…

Exciting Updates From Stanford HAI’s Seventh Annual AI Index Report

April 15, 2024

As the AI revolution marches on, it is vital to continually reassess how this technology is reshaping our world. To that end, researchers at Stanford’s Instit Read more…

Intel’s Vision Advantage: Chips Are Available Off-the-Shelf

April 11, 2024

The chip market is facing a crisis: chip development is now concentrated in the hands of the few. A confluence of events this week reminded us how few chips Read more…

The VC View: Quantonation’s Deep Dive into Funding Quantum Start-ups

April 11, 2024

Yesterday Quantonation — which promotes itself as a one-of-a-kind venture capital (VC) company specializing in quantum science and deep physics  — announce Read more…

Nvidia’s GTC Is the New Intel IDF

April 9, 2024

After many years, Nvidia's GPU Technology Conference (GTC) was back in person and has become the conference for those who care about semiconductors and AI. I Read more…

Google Announces Homegrown ARM-based CPUs 

April 9, 2024

Google sprang a surprise at the ongoing Google Next Cloud conference by introducing its own ARM-based CPU called Axion, which will be offered to customers in it Read more…

Computational Chemistry Needs To Be Sustainable, Too

April 8, 2024

A diverse group of computational chemists is encouraging the research community to embrace a sustainable software ecosystem. That's the message behind a recent Read more…

Hyperion Research: Eleven HPC Predictions for 2024

April 4, 2024

HPCwire is happy to announce a new series with Hyperion Research  - a fact-based market research firm focusing on the HPC market. In addition to providing mark 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…

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…

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…

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…

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…

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…

Leading Solution Providers

Contributors

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…

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…

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…

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…

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…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

Intel’s Xeon General Manager Talks about Server Chips 

January 2, 2024

Intel is talking data-center growth and is done digging graves for its dead enterprise products, including GPUs, storage, and networking products, which fell to Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire