Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “SGI”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3

Parallelization of NAS Benchmarks for Shared Memory Multiprocessors

This paper presents our experiences of parallelizing the sequential implementation of NAS benchmarks using compiler directives on SGI Origin2000 distributed shared memory (DSM) system. Porting existing applications to new high performance parallel and distributed computing platforms is a challenging task. Ideally, a user develops a sequential version of the application, leaving the task of porting to new generations of high performance computing systems to parallelization tools and compilers. Due to the simplicity of programming shared-memory multiprocessors, compiler developers have provided various facilities to allow the users to exploit parallelism. Native compilers on SGI Origin2000 support multiprocessing directives to allow users to exploit loop-level parallelism in their programs. Additionally, supporting tools can accomplish this process automatically and present the results of parallelization to the users. We experimented with these compiler directives and supporting tools by parallelizing sequential implementation of NAS benchmarks. Results reported in this paper indicate that with minimal effort, the performance gain is comparable with the hand-parallelized, carefully optimized, message-passing implementations of the same benchmarks.

Waheed, Abdul↗

Trace-Driven Debugging of Message Passing Programs

In this paper we report on features added to a parallel debugger to simplify the debugging of parallel message passing programs. These features include replay, setting consistent breakpoints based on interprocess event causality, a parallel undo operation, and communication supervision. These features all use trace information collected during the execution of the program being debugged. We used a number of different instrumentation techniques to collect traces. We also implemented trace displays using two different trace visualization systems. The implementation was tested on an SGI Power Challenge cluster and a network of SGI workstations.

Frumkin, Michael↗

Scaling NASA Applications to 1024 CPUs on Origin 3K

The long and highly successful joint SGI-NASA research effort in ever larger SSI systems was to a large degree the result of the successful development of the MLP scalable parallel programming paradigm developed at ARC: 1) MLP scaling in real production codes justified ever larger systems at NAS; 2) MLP scaling on 256p Origin 2000 gave SGl impetus to productize 256p; 3) MLP scaling on 512 gave SGI courage to build 1024p O3K; and 4) History of MLP success resulted in IBM Star Cluster based MLP effort.

Taft, Jim↗

Parallel 3D Mortar Element Method for Adaptive Nonconforming Meshes

High order methods are frequently used in computational simulation for their high accuracy. An efficient way to avoid unnecessary computation in smooth regions of the solution is to use adaptive meshes which employ fine grids only in areas where they are needed. Nonconforming spectral elements allow the grid to be flexibly adjusted to satisfy the computational accuracy requirements. The method is suitable for computational simulations of unsteady problems with very disparate length scales or unsteady moving features, such as heat transfer, fluid dynamics or flame combustion. In this work, we select the Mark Element Method (MEM) to handle the non-conforming interfaces between elements. A new technique is introduced to efficiently implement MEM in 3-D nonconforming meshes. By introducing an "intermediate mortar", the proposed method decomposes the projection between 3-D elements and mortars into two steps. In each step, projection matrices derived in 2-D are used. The two-step method avoids explicitly forming/deriving large projection matrices for 3-D meshes, and also helps to simplify the implementation. This new technique can be used for both h- and p-type adaptation. This method is applied to an unsteady 3-D moving heat source problem. With our new MEM implementation, mesh adaptation is able to efficiently refine the grid near the heat source and coarsen the grid once the heat source passes. The savings in computational work resulting from the dynamic mesh adaptation is demonstrated by the reduction of the the number of elements used and CPU time spent. MEM and mesh adaptation, respectively, bring irregularity and dynamics to the computer memory access pattern. Hence, they provide a good way to gauge the performance of computer systems when running scientific applications whose memory access patterns are irregular and unpredictable. We select a 3-D moving heat source problem as the Unstructured Adaptive (UA) grid benchmark, a new component of the NAS Parallel Benchmarks (NPB). In this paper, we present some interesting performance results of ow OpenMP parallel implementation on different architectures such as the SGI Origin2000, SGI Altix, and Cray MTA-2.

Feng, Huiyu↗

HEC Applications on Columbia Project

NASA's Columbia system consists of a cluster of twenty 512 processor SGI Altix systems. Each of these systems is 3 TFLOP/s in peak performance - approximately the same as the entire compute capability at NAS just one year ago. Each 512p system is a single system image machine with one Linunx O5, one high performance file system, and one globally shared memory. The NAS Terascale Applications Group (TAG) is chartered to assist in scaling NASA's mission critical codes to at least 512p in order to significantly improve emergency response during flight operations, as well as provide significant improvements in the codes. and rate of scientific discovery across the scientifc disciplines within NASA's Missions. Recent accomplishments are 4x improvements to codes in the ocean modeling community, 10x performance improvements in a number of computational fluid dynamics codes used in aero-vehicle design, and 5x improvements in a number of space science codes dealing in extreme physics. The TAG group will continue its scaling work to 2048p and beyond (10240 cpus) as the Columbia system becomes fully operational and the upgrades to the SGI NUMAlink memory fabric are in place. The NUMlink uprades dramatically improve system scalability for a single application. These upgrades will allow a number of codes to execute faster at higher fidelity than ever before on any other system, thus increasing the rate of scientific discovery even further

Taft, Jim↗

Glenn Heat Transfer Simulation and Solver Graphical User Interface: Development and Testing

In the Tui ine Branch of the Turbomachinery and Propulsion Systems Division, researching and developing efficient turbine aerothermodynamics technologies is the main objective. Creating effective turbines for jet engines is a process which, if based purely on physical experimental testing, would be extremely expensive. It is for this reason, and also for the reasons of speed and ease, that the Turbine Branch spends a large amount of effort working with simulations of turbines. Specifically, they focus their work on two main fields: Computational Field Dynamics (CFD), and Experimental data analysis. The experimental field involves comparing experimental results to simulated results, whereas the CFD field involves running these simulations. The simulations are applied to aerodynamics and heat transfer cases, for both steady and unsteady flow conditions. By and large this work is applied to the domain of flow and heat transfer in axial turbines. The main application used to run these heat flow simulations is GlennHT. This program, recently rewritten in FORTRAN 90, allows the user to input a job file which specifies all the necessary parameters needed to simulate flow through a user-defined grid. There are several other executables used as well, ranging in application from converting grid files to and from particular formats, to merging blocks in a connectivity file, to converting connectivity files to a GlennHT compatible format. All of these executables are run from the command line in a terminal; some of them have interactive prompts where the user must specify the files to be manipulated after the program starts, while others take all of their parameters from the command line. With this amount of variation comes a good deal of commands and formats to memorize, which can cause slower and less efficient work, as users may forget how to execute a certain program, or not remember the pathnames of the files they wish to use. Two years ago, steps were made to expedite this process with a graphical user interface (GUI) that combines the functionality of all the executables along with adding some new functionality, such as residuals graphing and boundary conditions creation. Upon my beginning here at Glenn, many parts of the GUI, which was developed in Java, were nonfunctional. There were also issues with cross-platforming, as systems in the branch were transitioning from Silicon Graphics (SGI) machines to Linux machines. My goals this summer are to finish the parts of the GUI that are not yet completed, fix parts that did not work correctly, expand the functionality to include other useful features, such as grid surface highlighting, and make the system compatible with both Linux and SGI. I will also be heavily testing the system and providing sufficient documentation on how to use the GUI, as no such documentation existed previously.

Kardamis, Joseph R.↗

Parallel Rendering of Large Time-Varying Volume Data

Interactive visualization of large time-varying 3D volume datasets has been and still is a great challenge to the modem computational world. It stretches the limits of the memory capacity, the disk space, the network bandwidth and the CPU speed of a conventional computer. In this SURF project, we propose to develop a parallel volume rendering program on SGI's Prism, a cluster computer equipped with state-of-the-art graphic hardware. The proposed program combines both parallel computing and hardware rendering in order to achieve an interactive rendering rate. We use 3D texture mapping and a hardware shader to implement 3D volume rendering on each workstation. We use SGI's VisServer to enable remote rendering using Prism's graphic hardware. And last, we will integrate this new program with ParVox, a parallel distributed visualization system developed at JPL. At the end of the project, we Will demonstrate remote interactive visualization using this new hardware volume renderer on JPL's Prism System using a time-varying dataset from selected JPL applications.

Garbutt, Alexander E.↗

Performance Evaluation of Supercomputers using HPCC and IMB Benchmarks

The HPC Challenge (HPCC) benchmark suite and the Intel MPI Benchmark (IMB) are used to compare and evaluate the combined performance of processor, memory subsystem and interconnect fabric of five leading supercomputers - SGI Altix BX2, Cray XI, Cray Opteron Cluster, Dell Xeon cluster, and NEC SX-8. These five systems use five different networks (SGI NUMALINK4, Cray network, Myrinet, InfiniBand, and NEC IXS). The complete set of HPCC benchmarks are run on each of these systems. Additionally, we present Intel MPI Benchmarks (IMB) results to study the performance of 11 MPI communication functions on these systems.

Saini, Subhash↗

Impact on TRMM Products of Conversion to Linux

In June 2008, TRMM data processing will be assumed by the Precipitation Processing System (PPS). This change will also mean a change in the hardware production environment from an SGI 32 bit IRIX processing environment to a Linux (Beowulf) 64 bit processing environment. This change of platform and operating system addressing (32 to 64) has some influence on data values in the TRMM data products. This paper will describe the transition architecture and scheduling. It will also provide an analysis of what the nature of the product differences will be. It will demonstrate that the differences are not scientifically significant and are generally not visible. However, they are not always identical with those which the SGI would produce.

Stocker, Erich Franz↗

Convective Transport of Very-short-lived Bromocarbons to the Stratosphere

We use the NASA GEOS Chemistry Climate Model (GEOSCCM) to quantify the contribution of two most important brominated very short-lived substances (VSLS), bromoform (CHBr3) and dibromomethane (CH2Br2), to stratospheric bromine and its sensitivity to convection strength. Model simulations suggest that the most active transport of VSLS from the marine boundary layer through the tropopause occurs over the tropical Indian Ocean, the Western Pacific warm pool, and off the Pacific coast of Mexico. Together, convective lofting of CHBr3 and CH2Br2 and their degradation products supplies ∼8 ppt total bromine to the base of the Tropical Tropopause Layer (TTL, ∼150 hPa), similar to the amount of VSLS organic bromine available in the marine boundary layer (∼7.8-8.4 ppt) in the above active convective lofting regions. Of the total ∼8 ppt VSLS-originated bromine that enters the base of TTL at ∼150 hPa, half is in the form of source gas injection (SGI) and half as product gas injection (PGI). Only a small portion (< 10%) the VSLS-originated bromine is removed via wet scavenging in the TTL before reaching the lower stratosphere. On global and annual average, CHBr3 and CH2Br2, together, contribute ∼7.7 pptv to the present-day inorganic bromine in the stratosphere. However, varying model deep convection strength between maximum and minimum convection conditions can introduce a ∼2.6 pptv uncertainty in the contribution of VSLS to inorganic bromine in the stratosphere (BryVSLS). Contrary to the conventional wisdom, minimum convection condition leads to a larger BryVSLS as the reduced scavenging in soluble product gases, thus a significant increase in PGI (2-3 ppt), greatly exceeds the relative minor decrease in SGI (a few 10ths ppt.

bromine↗

Implementing a Gaussian Process Learning Algorithm in Mixed Parallel Environment

In this paper, we present a scalability analysis of a parallel Gaussian process training algorithm to simultaneously analyze a massive number of time series. We study three different parallel implementations: using threads, MPI, and a hybrid implementation using threads and MPI. We compare the scalability for the multi-threaded implementation on three different hardware platforms: a Mac desktop with two quad-core Intel Xeon processors (16 virtual cores), a Linux cluster node with four quad-core 2.3 GHz AMD Opteron processors, and SGI Altix ICE 8200 cluster node with two quad-core Intel Xeon processors (16 virtual cores). We also study the scalability of the MPI based and the hybrid MPI and thread based implementations on the SGI cluster with 128 nodes (2048 cores). Experimental results show that the hybrid implementation scales better than the multi-threaded and MPI based implementations. The hybrid implementation, using 1536 cores, can analyze a remote sensing data set with over 4 million time series in nearly 5 seconds while the serial algorithm takes nearly 12 hours to process the same data set.

99 GENERAL AND MISCELLANEOUS↗

Snake River Plain Geothermal Play Fairway Analysis Project Active Source Seismic Data

This archive contains seismic shot field records for 10 profiles located in Camas Prairie, Idaho. The eight numbered .sgy files were acquired using a seismic land streamer system with an accelerated weight drop source and 72 geophones. These 10-Hz geophones were mounted on base plates and dragged behind the seismic source. Shots were acquired every 4 meters along the length of lines 500West, 550 West, 600West, 700West, 800West, 900West, 200South and 200North. The objective was to map stratigraphy and structures related to geothermal fluid flow in the upper few hundred meters. A readme file is included with descriptions of individual files. The lines names refer to to roads which are numbered relative to the distance from the county seat (the town of Fairfield) along the the main highways. For example, 500 West implies that this north-south street crosses the main road 5 miles to the west of town. The included geologic, topographic, and aerial maps show the labeled seismic lines, while the regional map shows only the line geometry and regional faulting.

15 GEOTHERMAL ENERGY↗

EGS Collab Testbed 1: Baseline Cross-well Seismic

As part of the geophysical characterization suite for the first EGS Collab tesbed, here are the baseline cross-well seismic data and resultant models. The campaign seismic data have been organized, concatenated with geometry and compressional (P-) & and shear (S-) wave picks, and submitted as SGY files. P-wave data were collected and analyzed in both 2D and 3D, while S-wave data were collected and analyzed in 2D only. Inversion models are provided as point volumes; the volumes have been culled to include only the points within source/receiver array coverage. The full models space volumes are also included, if relevant. An AGU 2018 poster by Linneman et al. is included that provides visualizations/descriptions of the cross-well seismic characterization method, elastic moduli calculations, and images of model inversion results.

15 GEOTHERMAL ENERGY↗

INL's Data Center

ICE STORM is a supercomputer procured by INL from a well-known computer vendor, SGI. ICESTORM is rated as No. 64 on the list of 500 fastest supercomputers today. It has 2,048 processors all working together.

97 MATHEMATICS AND COMPUTING↗

Graph 500 in OpenSHMEM, In: OpenSHMEM 2014: OpenSHMEM and Related Technologies. Experiences, Implementations, and Technologies

This document describes the effort to implement the Graph 500 benchmark using OpenSHMEM based on the MPI-2 one-side version. The Graph 500 benchmark performs a breadth-first search in parallel on a large randomly generated undirected graph and can be implemented using basic MPI-1 and MPI-2 one-sided communication. Graph 500 requires atomic bit-wise operations on unsigned long integers but neither atomic bit-wise operations nor OpenSHMEM for unsigned long are available in OpenSHEM. Such needed bit-wise atomic operations and support for unsigned long are implemented using atomic condition swap (CSWAP) on signed long integers. Preliminary results on comparing the OpenSHMEM and MPI-2 one-sided implementations on a Silicon Graphics Incorporated (SGI) cluster and the Cray XK7 are presented.

Computer Science↗

OpenSHMEM Implementation of HPCG Benchmark, In: OpenSHMEM 2016: OpenSHMEM and Related Technologies. Enhancing OpenSHMEM for Hybrid Environments

We describe the effort to implement the HPCG benchmark using OpenSHMEM and MPI one-sided communication. Unlike the High Performance LINPACK (HPL) benchmark that places emphasis on large dense matrix computations, the HPCG benchmark is dominated by sparse operations such as sparse matrix-vector product, sparse matrix triangular solve, and long vector operations. The MPI one-sided implementation is developed using the one-sided OpenSHMEM implementation. Preliminary results comparing the original MPI, OpenSHMEM, and MPI one-sided implementations on an SGI cluster, Cray XK7 and Cray XC30 are presented. The results suggest the MPI, OpenSHMEM, and MPI one-sided implementations all obtain similar overall performance but the MPI one-sided implementation seems to slightly increase the run time for multigrid preconditioning in HPCG on the Cray XK7 and Cray XC30.

Computer Science↗

Protein kinase C alpha-mediated phosphorylation of PIM-1L promotes the survival and proliferation of acute myeloid leukemia cells

FMS-like tyrosine kinase 3 (FLT3)-internal tandem duplication (ITD) is a constitutively active mutant of FLT3 and causes 20%–30% of acute myeloid leukemia (AML) cases. FLT3-ITD upregulates the proviral integration site for Moloney murine leukemia virus 1 (PIM-1) expression and promotes the proliferation of AML cells. In this study, we investigated the role of protein kinase C (PKC)-mediated phosphorylation on the expression and function of PIM-1L. Drug screening in leukemia cell lines revealed that sotrastaurin (a PKC inhibitor) suppressed the proliferation of the FLT3-ITD-positive AML cell line MV4-11 but not of K562, HL60, or KG-1a cells, similar to SGI-1776 (a PIM-1/FLT3 inhibitor) and quizartinib (an FLT3 inhibitor). Sotrastaurin decreased the expression of pro-survival protein myeloid cell leukemia (MCL-1) and the phosphorylation of eukaryotic initiation factor 4E-binding protein 1 (4E-BP1), both of which are downstream effectors of PIM-1. PKCα directly phosphorylated Ser65 of PIM-1L, which is a long isoform of PIM-1. The PKCα-mediated phosphorylation stabilized PIM-1L. The phosphorylation-mimicked mutant, PIM-1L-S65D, was more stable and showed higher kinase activity than PIM-1L-S65A. Expression of PIM-1L-wildtype or -S65D reduced sotrastaurin-mediated apoptosis and growth inhibition in MV4-11 transfectants. These results suggest that PKCα directly upregulates PIM-1L, resulting in promotion of the survival and proliferation of AML cells.

60 APPLIED LIFE SCIENCES↗

Feedback-directed page placement for ccNUMA via hardware-generated memory traces

Non-uniform memory architectures with cache coherence (ccNUMA) are becoming increasingly common, not just for large-scale high performance platforms but also in the context of multi-core architectures. Under ccNUMA, data placement may influence overall application performance significantly as references resolved locally to a processor/core impose lower latencies than remote ones. This work develops a novel hardware-assisted page placement paradigm based on automated tracing of the memory references made by application threads. Two placement schemes, modeling both singlelevel and multi-level latencies, allocate pages near processors that most frequently access that memory page. These schemes leverage performance monitoring capabilities of contemporary microprocessors to efficiently extract an approximate trace of memory accesses. This information is used to decide page affinity, i.e., the node to which the page is bound. The method operates entirely in user space, is widely automated, and handles not only static but also dynamic memory allocation. Experiments show that this method, although based on lossy tracing, can efficiently and effectively improve page placement, leading to an average wall-clock execution time saving of over 20% for the tested benchmarks on the SGI Altix with a 2x remote access penalty and 12% on AMD Opterons with a 1.3–2.0x access penalty. This is accompanied by a one-time tracing overhead of 2.7% over the overall original program wallclock time.

Computer Science↗