NCAR's Advanced Research Computing System (ARCS)
|
The following information is excerpted from SCD's FY20002 Annual Scientific Report.
The production supercomputer environment managed by SCD for NCAR has evolved over the decades. During the last 18 years, SCD has brought NCAR's science into the world of multiprocessing supercomputers. Prior to the introduction of the four-CPU Cray X-MP in October 1986, all modeling was performed with serial codes. Since then, the focus has been on redeveloping codes to harness the power of multiple CPUs in a single system, and most recently, in multiple systems.
SCD's computer room is now dominated by Distributed Shared Memory (DSM) systems, most notably the IBM SP system blackforest and the larger and more powerful IBM Cluster 1600 system bluesky.
Early in FY2002, phase I of the Advanced Research Computing System (ARCS) was delivered. This more than doubled the size of blackforest, to 1,308 processors. Shortly after this upgrade, a new IBM p690 system with 16 POWER4 processors was delivered. This machine was used for testing and system development in preparation for a large IBM Cluster 1600 system scheduled for delivery in October 2002. Also during FY2002, four systems were decommissioned and two new systems installed. In addition, there were major system software upgrades performed on all major supercomputers.
SCD Director Al Kellie says, "When we designed the ARCS RFP, we wanted a production-level, high-performance computing system that offered both capability and capacity computing. But more than that, we wanted to provide an upwardly compatible system architecture and user environment, and a stable environment for software engineering and development."
Thus, existing blackforest users found the same operating system, the same node configuration, the same batch system (LoadLeveler), and the same industry-standard debugger (TotalView). Usage of the new system continued to be split 50/50 between Community Computing and the Climate Simulation Laboratory (CSL), as it was before the upgrade. But whereas blackforest had become saturated in FY2001, the upgraded system offered plenty of capacity.
To prepare NCAR's computer room for the new ARCS equipment arriving in FY2002, SCD staff started work in early FY2001 to specify, procure, and install a new electrical power distribution system to support the next three years of the ARCS contract. This forms the cornerstone of a solid infrastructure for SCD to continue providing reliable, production-oriented services and equipment as tools for science.
The upgraded blackforest system went into production in December 2001. The National Science Foundation purchased the machine for use at NCAR to advance a wide range of research topics in the agency's ten-year plan for the geosciences.
|
