June 09, 2008
Sure, the race for the White House gets most of the coverage, but HP made its own campaign announcement recently, pledging to reinforce its commitment to constituents in datacenters everywhere. As part of its new "scalable computing initiative," the company has formed a business unit to serve high-performance, high-demand computing operations. HP promised new products and services to solve the problems particular to large-scale scale-out environments such as clouds and grids, while also delivering the real pork: lower infrastructure costs.
To HP's credit, it rolled out a new product aimed right at some top concerns of datacenter managers. The new ProLiant BL2x220c blade server is the first to pack two independent servers in one blade. Obviously, this design innovation by itself means a savings in floor space and real estate costs, but for some managers in fast-growing businesses -- the ones who've figured out they're going to need space for 200 new servers and the last open corner will hold only 100 -- the new design could be an unexpected solution. HP says its new blade also draws less electric power than "similar competing offerings." With two servers per slot, a single 42U rack can be stuffed with as many as 128 servers, 1,024 CPU cores, and 2TB of memory.
"Customers have told us that their computing environments are running out of floor space and using too much energy -- and therefore money -- to power their applications," Ed Turkel, manager of HPC product marketing for the new Scalable Computing & Infrastructure organization at HP, told GRIDtoday. "As a vital part of the business, IT professionals are also looking to cut costs associated with their facilities and infrastructure. Our new scale-out offerings ... are designed to take the best advantage of the space and energy available in the datacenter. The BL2x22Oc doubles compute power but takes up half the space of a traditional server blade with 68 percent better performance per watt compared to the Dell PowerEdge M600."
The new ProLiant is built around Intel's Xeon 5400 Series Quad Core CPUs or Xeon 5200 Series Dual Core CPUs (up to 2 per server node). HP says a maxed-out 42U rack would deliver 12.3 teraflops while requiring only 8 square feet of floor space. Gigabit Ethernet connectivity and HP's Integrated Lights-Out 2 remote management and system monitoring software are standard.
As part of its product portfolio, the new HP scalable computing unit also is offering the recently introduced StorageWorks 9100 Extreme Data Storage System, a highly scalable unit designed to simplify management of multiple petabytes -- and a perfect complement to the blade system. "The BL2x220c and StorageWorks 9100 storage system are ideal for scale-out environments that require massive computing power, but aren't limited to cloud and high performance computing customers," Turkel says. Their benefits expand to customers in online social communities, financial modeling, digital animation, online gaming, Web search, oil and gas, electronic design automation and biotechnology research, he says.
Smaller Footprints Pay Off
Swedish Medical Center, a three-hospital health care provider in Seattle, has been beta testing the new blade server for about six months. The center's IT department is using the BL2x220c to run a farm of Citrix servers used for the Epic electronic medical record (EMR) system, says Mike Criss, manager of infrastructure and engineering. "We have it in production, and it's been great," he says.
The denser blades are arriving at a good time for Swedish. "We are at about 75 percent capacity in our datacenter," Criss says. "If we continued to add servers at the same rate in the same form factor, we'd run out of space in a year easily. We have some big growth coming in terms of the EMR system, so we're excited by the design and the density of the new 2x220c blade. It will let us grow without having to make a massive expansion to the datacenter."
Criss's only concern with the two-in-one design: "If one server goes out, you have to take them both offline when you pull out the chassis."
Another user is WETA Digital, a New Zealand-based digital imaging company that did special effects rendering for "Lord of the Rings" and "King Kong." Animators are globally dispersed, so WETA needed a high-performance system that could handle heavy computational demands and a high volume of extremely large files in real time. Turkel says that by implementing the new BL2x220c, WETA was able to add a significantly higher number of processors into a confined physical footprint while still managing within tight power consumption and cooling requirements.
Along with the new hardware, HP's scalability group will be offering facilities planning services to help customers design, upgrade, and retrofit datacenters to better utilize space and be more efficient, Turkel says. "These 'Data Center Transformation services' are technology tools and strategies to deliver top datacenter initiatives such as automation, energy efficiency, virtualization, consolidation and business continuity."
"Coming out off the Closet"
Of course, HP is not new to the business of serving customers with big scale-up needs. The new Scalable Computing & Infrastructure organization combines resources from two groups that have been working in those environments, the High-Performance Computing and Scalable Datacenter Infrastructure teams. That's why HP's recent announcement of the initiative was "more of a coming out of the closet," says James Staten, a principal analyst who covers server infrastructures and datacenters for Forrester Research. "It's HP saying, 'Hey, for the last three years we've been building products for customers in these markets.'"
HP's Turkel says the new organization is significant because it "will drive development of offerings that address the unique needs of scale-out computing customers resulting from the convergence of high-performance computing and Web 2.0, and the emergence of cloud-based business models."
"We have cloud domain expertise and dedicated HP resources who have been servicing these customers for years and understand the requirements of scale-out," he says. "HP is servicing scale-out and cloud-like computing environments for customers with tens of thousands of servers and tens of petabytes of storage installed around the world."
According to Forrester analyst Staten, however, HP's new organization is not nearly as significant as the new blade server.
"The real news here is the double-decker server," Staten says. "That's innovative design. HP has done something clever, and it's the kind of thing the Web giants find interesting. They're interested in compute density. And the big datacenters, they're definitely concerned about powered and cooled floor space. You get to big labs like the NCSA [National Center for Supercomputer Applications], they're not going to build a new datacenter. The space they have is the space they have. So HP's two-in-one blade server would be a big step forward for them."
May 16, 2013 |
When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
May 10, 2013 |
Australian visual effects company, Animal Logic, is considering a move to the public cloud.
May 10, 2013 |
Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
May 08, 2013 |
For engineers looking to leverage high-performance computing, the accessibility of a cloud-based approach is a powerful draw, but there are costs that may not be readily apparent.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/02/2012 | AMD | Developers today are just beginning to explore the potential of heterogeneous computing, but the potential for this new paradigm is huge. This brief article reviews how the technology might impact a range of application development areas, including client experiences and cloud-based data management. As platforms like OpenCL continue to evolve, the benefits of heterogeneous computing will become even more accessible. Use this quick article to jump-start your own thinking on heterogeneous computing.