During Supercomputing '11 in Seattle, Washington, NOAA will be presenting their MOAB application during a Birds of a Feather on November 17th at 12:15 in room TCC 305. The Moab decision engine is the workload management software for Gaea, NOAA's new leadership class supercomputer, and as the standard for providing HPC Grid functionality to all NOAA supercomputers. With Moab, NOAA gains a robust management infrastructure for compute jobs that unifies HPC resources across large geographic divides and maximizes job throughput and CPU utilization to deliver on the project's overall goal of developing better models for predicting climate variability and change.
In choosing a workload manager, one of NOAA's primary considerations was location-aware scheduling. NOAA's Geophysical Fluid Dynamics Laboratory (GFDL), located in Princeton, New Jersey supports their local researchers as well as other NOAA researchers across the country. Gaea is physically located at ORNL in Tennessee. The disparate locations of users and system(s), current and future, create challenges in networking, data transfer, and job submissions. Moab solves the job submission problem by allowing a local instance of Moab to be installed in New Jersey where users can interact with the system, manipulate their data sets, and analyze their results. Moab then communicates with, and migrates workload jobs and data between GFDL and the instance of Moab running on Gaea in Tennessee. This model can grow organically as new users and compute resources come on-line.
Moab is unique among workload managers as it can run on multiple resource managers. This capability is a crucial component to NOAA's goal of delivering a unified Grid. On Gaea, NOAA plans to use Moab with TORQUE Resource Manager, a PBS-based open-source resource manager that is maintained and supported by Adaptive Computing.
"NOAA's mission is to understand and predict changes in the Earth's environment and we rely on supercomputing technologies like Moab to support the data-intensive research of our scientists", stated Joseph Klimavicz, chief information officer and director of high performance computing and communications at NOAA. "We look forward to working with a well-established HPC software provider such as Adaptive Computing and are confident in the product's capabilities."
"We selected Adaptive Computing for NOAA's mission-critical deployment based on the company's proven Moab technology and its unique, location-aware functionality", stated Steven Baxter, program manager at CSC.
NOAA is currently licensed to run Moab at three other HPC sites, including Boulder, Colorado and the $27.6 million supercomputing centre in Fairmont, West Virginia. NOAA's long-term plan is to link the sites under a single HPC Grid for global job submission and a single point of reporting.
"We are honoured to play a critical role in supporting NOAAs ground-breaking climate research", stated Robert Clyde, CEO of Adaptive Computing. "As HPC systems grow more complex, flexibility is a key component for any resource management solution. The latest upgrades to the Moab and Viewpoint technology enable the type of flexibility required for next-generation supercomputers."
Funded through the American Recovery and Reinvestment Act of 2009, Gaea will serve as a dedicated high performance computing resource for NOAA and its extensive network of research partners. This system will enable scientists to leverage a significant increase in computing capacity to address some of the most pressing global climate change questions. Moab manages more Top500 CPUs than any other solution. It is experienced in managing large numbers of users in complex research environments, while simultaneously optimizing the utilization of peta-flop-scale supercomputers.