What It's Like To....Build The World's Most Powerful Supercomputer

Deputy Associate Director of Computation at Lawrence Livermore National Lab, Michel McCoy does extraordinary work.

By Michel McCoy & Alice Dragoon
Wed, December 15, 2004

CIO — When I picked up the phone late in September, a very pleased and excited research leader from IBM was on the line, telling me that we'd just made history. Since February 2000, we'd been working together to build a new supercomputer for the Department of Energy, to be sited at Lawrence Livermore National Lab. IBM was calling to tell me that a small prototype of our Blue Gene/L supercomputer had just roared through 36.01 teraflops—more than 36 trillion calculations!—in one second. Testing just eight racks of what will be a 64-rack system, IBM scientists had beaten the previous world record of 35.86 teraflops set in 2002 by NEC's Earth Simulator in Japan.

By next June, Blue Gene will hit well over 200 teraflops.

We'd been worried we didn't have enough time to get anything running on the new machine with the new operating system and a new software stack. So reaching 70 percent of peak capacity at this point was extremely exciting. It was like the day my first daughter was born. I felt like we'd gotten a sign from the gods that we'd done at least a few things right. Then in November, we earned the number-one spot on the list of the world's fastest supercomputers when a 16-rack version of Blue Gene/L tested out at 70.72 teraflops. By next June we'll have all 64 racks installed and we expect to be able to hit well over 200 teraflops—the equivalent of everyone on earth performing 35,000 calculations per second.

We still have to harden the technology. It's not superstabilized. But we're talking about something utterly new and really exciting. We're enabling scientists in ways they didn't believe possible. In the past, studying the transition to turbulence in fluids required using approximations; more detailed, atomistic calculations were prohibitively expensive. I was talking recently with a scientist about Blue Gene's capacity and he got all excited, saying, Oh my God, I can run calculations on 50 billion atoms instead of just 1 billion!

To be working on a leading-edge project like this is one of the things that makes this life worthwhile. My life is pretty ordinary overall but to be part of something that's unique and has such promise to be valuable to the laboratory and to the nation is extraordinarily special.

—As told to Alice Dragoon

This paper covers power utilization, intelligent power management and industry best practices for energy efficiency. Extreme Networks® takes a lifecycle approach to power efficiency, management and recycling, offering savings to our customers and promoting a greener world.
With increasing data growth, comes increased need for data security.  The existing DLP model, with a focus on compliance/enforcement is not sufficient as the data discovery and classification capabilities are not granular enough.  Read this paper to find how you can efficiently and accurately manage your risk by rapidly inventorying and classifying your data and then developing remediation workflows that support business needs. 
This paper breaks down attack sources into four categories: external, malicious insiders, accidental insiders, and unknown.
The rapid growth of data and technology is creating challenges for organizations as this digital data is considered to be business communications and must be preserved according the same industry-specific regulations governing the retention and discovery of emails and more traditional forms of electronic communications. This paper examines the role that Data Loss Prevention ("DLP") technology can play in helping organizations address the challenges of locating information in response to electronic discovery.
This research, conducted by the Ponemon Institute, focuses on issues relating to the use of data protection solutions such as endpoint encryption and data loss prevention within the workplace.
This report, by Jon Oltsik from Enterprise Strategy Group, examines the need for a new business-centric approach to DLP in order to align business and security requirements.
Have you been looking to hear about customer's experiences with the new VMware vCenter Site Recovery Manager product? View this webcast to learn about VMware customer, Navicure, and their experiences testing and evaluating the recovery manager, their progress in implementing it in their environment and their advice other customers considering using vCenter.
Virtualizing business-critical applications is an essential step in your journey to the cloud. Microsoft SQL Server, Exchange and SharePoint, and Oracle applications, are often the backbone of business IT. The benefits of virtualizing these applications extend far beyond mere consolidation. Understanding how VMware improves quality of service and agility while reducing costs will help you make the case for taking virtualization to the next level in your company.
Applications are changing - they're increasingly web-oriented, global in nature and run from multiple device types. Additionally, the volume of data is growing exponentially every year. How do you ensure your applications have fast, accurate, up-to-date information in this new world? Modern applications are data-intensive; delivering data the old way using monolithic databases isn't working. What's needed is a modern approach to data. One that scales-out as needed and delivers predictable high performance, but without sacrificing data consistency or integrity.
Real-time, global data updates have become a critical business requirement for financial-services firms. Overnight or hourly batch jobs can cause erroneous results and missed opportunities. New regulatory requirements dictate real-time reporting of liquidity; traders want access to real-time market and risk positions; and the time windows for relevancy of cross-selling and marketing opportunities are getting shorter. To deal with these issues and new requirements, firms need to be able to react quickly to changes in data. Quick reactions require near-instant access to data, risk analysis and deeper computational analysis for effective decision making. View this webcast to learn how to achieve real-time awareness by managing ever-increasing data volumes and transaction rates.
This video webcast is designed to help those with little to no virtualization experience understand why virtualization and VMware are so important to driving down both capital and operational costs. The session will start with the introduction of the key concepts and technologies of virtualization, introduce the vSphere Hypervisor, and build up to an overview of VMware vSphere® 5, the world's most robust and complete virtualization platform. This session will also discuss new solutions such as the vSphere Storage Appliance and VMware GO that are making it easier than ever before to get started with virtualization.
Big Data-it has the potential of transforming a business. In the case of Klout, a social networking analytics site, big data is the heart of the business. Klout processes and analyzes billions of user data signals every day-from Facebook, Twitter, LinkedIn, blogs and more. How do they do it? Gain valuable insights from David Mariani, vice president of engineering for Klout.
Newsletter Sign-Up »

Receive the latest news test, reviews and trends on your favorite technology topics

Choose a newsletter
  1. View all Newsletters | Privacy Policy
Resource Center