Joel O. Stevenson

Toward Rapid Understanding of Production HPC Applications and Systems

A detailed understanding of HPC applications resource needs and their complex interactions with each other and HPC platform resources is critical to achieving scalability and performance. Such understanding has been difficult to achieve because typical application profiling tools do not capture the behaviors of codes under the potentially wide spectrum of actual production conditions and because typical monitoring tools do not capture system resource usage information with high enough fidelity to gain sufficient insight into application performance and demands. In this paper we present both system and application profiling results based on data obtained through synchronized system wide monitoring on a production HPC cluster at Sandia National Laboratories (SNL). We demonstrate analytic and visualization techniques that we are using to characterize application and system resource usage under production conditions for better understanding of application resource needs. Our goals are to improve application performance (through understanding application-to-resource mapping and system throughput) and to ensure that future system capabilities match their intended workloads.

Supercomputer and Cluster Performance Modeling and Analysis Efforts: 2004-2006

This report describes efforts by the Performance Modeling and Analysis Team to investigate performance characteristics of Sandias engineering and scientific applications on the ASC capability and advanced architecture supercomputers, and Sandias capacity Linux clusters. Efforts to model various aspects of these computers are also discussed. The goals of these efforts are to quantify and compare Sandias supercomputer and cluster performance characteristics; to reveal strengths and weaknesses in such systems; and to predict performance characteristics of, and provide guidelines for, future acquisitions and follow-on systems. Described herein are the results obtained from running benchmarks and applications to extract performance characteristics and comparisons, as well as modeling efforts, obtained during the time period 2004-2006. The format of the report, with hypertext links to numerous additional documents, purposefully minimizes the document size needed to disseminate the extensive results from our research.

Process uniformity for plasma etchback and desmear in printed wiring board manufacturing

In the manufacture of printed wiring boards (PWB), plasma etchback and desmear processes facilitate the making of good mechanical and electrical bonds of copper inner layers to copper plating. Without sufficient plasma treatment, internal layer copper features receive inadequate polymer removal which results in circuit discontinuity during the plating process. Additionally, the plasma serves to roughen the polymer wall of drilled holes which improves copper adhesion. To ensure proper plasma treatment, careful adherence to strict production guidelines is essential. These guidelines include attention to several critical criteria in placement, pretreatment and treatment of the PWBs during the plasma process; process verification via post plasma testing; and careful process monitoring throughout. In this brief, some guidelines for process monitoring and control will be discussed. A description of a new plasma monitor utilizing optical emission spectroscopy (OES), developed cooperatively between Sandia National Laboratories, National Consortium for Manufacturing Sciences (NCMS) and Texas Instruments Inc., will be discussed along with possible benefits derived from in situ monitoring of plasma systems.

Preliminary Assessment of Tecplot Chorus for Analyzing Ensemble of CTH Simulations

The exploration of large parameter spaces in search of problem solution and uncertainty quantifcation produces very large ensembles of data. Processing ensemble data will continue to require more resources as simulation complexity and HPC platform throughput increase. More tools are needed to help provide rapid insight into these data sets to decrease manual processing time by the analyst and to increase knowledge the data can provide. One such tool is Tecplot Chorus, whose strengths are visualizing ensemble metadata and linked images. This report contains the analysis and conclusions from evaluating Tecplot Chorus with an example problem that is relevant to Sandia National Laboratories.