TAU Commander: An Intuitive Interface for the TAU Performance System
Explore the source record for details and available documents.
SEARCH · Engineering Papers
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.
Explore the source record for details and available documents.
Many research questions remain open with regard to improving reliability in exascale systems. Among others, statistics-based analysis has been used to find anomalies, to isolate root causes, and attempt to predict failures. But well-understood methods and best practices for collecting reliability data in a uniform way are still lacking, which impedes analysis. We report our experience with collecting these data from heterogeneous sources on a testbed cluster and present our data collection tool. This case illustrates the fact that reported metrics largely depend upon individual system configuration. We then investigate standards and specifications in manufacturing and desktop computing to identify concepts that may be useful for representing High Performance Computing (HPC) data and present a taxonomy that utilizes these concepts.
Building the next-generation of extreme-scale distributed systems will require overcoming several challenges related to system resilience. As the number of processors in these systems grow, the failure rate increases proportionally. One of the most common sources of failure in large-scale systems is memory. In this paper, we propose a novel runtime for transparently exploiting memory content similarity to improve system resilience by reducing the rate at which memory errors lead to node failure. We evaluate the viability of this approach by examining memory snapshots collected from eight high-performance computing (HPC) applications and two important HPC operating systems. Based on the characteristics of the similarity uncovered, we conclude that our proposed approach shows promise for addressing system resilience in large-scale systems.
High-performance computing architectures face nontrivial data processing challenges, as computational and I/O components further diverge in performance trajectories. For scientific data analysis in particular, methods based on generating heavyweight access acceleration structures, e.g. indexes, are becoming less feasible for ever-increasing dataset sizes. We present ALACRITY, demonstrating the effectiveness of a fused data and index encoding of scientific, floating-point data in generating lightweight data structures amenable to common types of queries used in scientific data analysis. We exploit the representation of floating-point values by extracting significant bytes, using the resulting unique values to bin the remaining data along fixed-precision boundaries. To optimize query processing, we use an inverted index, mapping each generated bin to a list of records contained within, allowing us to optimize query processing with attribute range constraints. Overall, the storage footprint for both index and data is shown to be below numerous configurations of bitmap indexing, while matching or outperforming query performance.
We advocate domain-specific virtual processors (DSVP) as a portability layer for expressing and executing domain-specific computational workloads on modern heterogeneous HPC architectures, with applications in quantum chemistry. Specifically, in this article we extend, generalize and better formalize the concept of a domain-specific virtual processor as applied to scientific high-performance computing. In particular, we introduce a system-wide recursive (hierarchical) hardware encapsulation mechanism into the DSVP architecture and specify a concrete microarchitectural design of an abstract DSVP from which specialized DSVP implementations can be derived for specific scientific domains. Furthermore, we demonstrate, an example of a domain-specific virtual processor specialized to numerical tensor algebra workloads, which is implemented in the ExaTENSOR library developed by the author with a primary focus on the quantum many-body computational workloads on large-scale GPU-accelerated HPC platforms.
I will discuss several processors: 1. The Cray proprietary processor used in the Cray X1; 2. The IBM Power 3 and Power 4 used in an IBM SP 3 and IBM SP 4 systems; 3. The Intel Itanium and Xeon, used in the SGI Altix systems and clusters respectively; 4. IBM System-on-a-Chip used in IBM BlueGene/L; 5. HP Alpha EV68 processor used in DOE ASCI Q cluster; 6. SPARC64 V processor, which is used in the Fujitsu PRIMEPOWER HPC2500; 7. An NEC proprietary processor, which is used in NEC SX-6/7; 8. Power 4+ processor, which is used in Hitachi SR11000; 9. NEC proprietary processor, which is used in Earth Simulator. The IBM POWER5 and Red Storm Computing Systems will also be discussed. The architectures of these processors will first be presented, followed by interconnection networks and a description of high-end computer systems based on these processors and networks. The performance of various hardware/programming model combinations will then be compared, based on latest NAS Parallel Benchmark results (MPI, OpenMP/HPF and hybrid (MPI + OpenMP). The tutorial will conclude with a discussion of general trends in the field of high performance computing, (quantum computing, DNA computing, cellular engineering, and neural networks).
The increasing volume of scientific data and the limited scalability and performance of storage systems are currently presenting a significant limitation for the productivity of the scientific workflows running on both high-performance computing (HPC) and cloud platforms. Clearly needed is better integration of storage systems and workflow engines to address this problem. This paper presents and evaluates a novel solution that leverages codesign principles for integrating Hercules—an in-memory data store—with a workflow management system. We consider four main aspects: workflow representation, task scheduling, task placement, and task termination. As a result, the experimental evaluation on both cloud and HPC systems demonstrates significant performance and scalability improvements over existing state-of-the-art approaches.
Performance analysis of GPU-accelerated systems requires a system-wide view that considers both CPU and GPU components. In this paper, we describe how to extend system-wide, sampling-based performance analysis methods to GPU-accelerated systems. Since current GPUs do not support sampling, our implementation required careful coordination of instrumentation-based performance data collection on GPUs with sampling-based methods employed on CPUs. In addition, we also introduce a novel technique for analyzing systemic idleness in CPU/GPU systems. We demonstrate the effectiveness of our techniques with application case studies on Titan and Keeneland. Some of the highlights of our case studies are: 1) we improved performance for LULESH 1.0 by 30%, 2) we identified a hardware performance problem on Keeneland, 3) we identified a scaling problem in LAMMPS derived from CUDA initialization, and 4) we identified a performance problem that is caused by GPU synchronization operations that suffer delays due to blocking system calls.
Multiple data sources are used in a variety of biosurveillance systems. With the advent of new technologies, globalization, high performance computing, and "big data" opportunities, there are seemingly unlimited potential data streams that could be useful in biosurveillance. Data streams have not been universally defined in either the literature or by specific biosurveillance systems. Furthermore, the definitions and framework that we have developed enable a characterization methodology that facilitates understanding of data streams and can be universally applicable for use in evaluating and understanding a wide range of biosurveillance activities- filling a gap recognized in both the public health and biosurveillance communities.
As CPU clock speeds have stagnated and high performance computers continue to have ever higher core counts, increased parallelism is needed to take advantage of these new architectures. Traditional serial time-marching schemes can be a significant bottleneck, as many types of simulations require large numbers of time-steps which must be computed sequentially. Parallel-in-time schemes, such as the Multigrid Reduction in Time (MGRIT) method, remedy this by parallelizing across time-steps and have shown promising results for parabolic problems. However, chaotic problems have proved more difficult, since chaotic initial value problems (IVPs) are inherently ill-conditioned. MGRIT relies on a hierarchy of successively coarser time-grids to iteratively correct the solution on the finest time-grid, but due to the nature of chaotic systems, small inaccuracies on the coarser levels can be greatly magnified and lead to poor coarse-grid corrections. Here we introduce a modified MGRIT algorithm based on an existing quadratically converging nonlinear extension to the multigrid Full Approximation Scheme (FAS), as well as a novel time-coarsening scheme. Together, these approaches better capture long-term chaotic behavior on coarse-grids and greatly improve convergence of MGRIT for chaotic IVPs. Further, we introduce a novel low-memory variant of the algorithm for solving chaotic PDEs with MGRIT which not only solves the IVP, but also provides estimates for the unstable Lyapunov vectors of the system. Finally, we provide supporting numerical results for the Lorenz system and demonstrate parallel speedup for the chaotic Kuramoto–Sivashinsky PDE over a significantly longer time-domain than in previous works.
The inverter-based resource (IBR) has become avery important component in the distribution system. The impacts on system transient stability introduced by high IBR penetration are not fully addressed because of the lack of high-fidelity models. The aggregate IBR model at the transmission level cannot precisely reproduce the dynamics of distributed IBR at the distribution system because of the oversimplification. In this paper, we will develop a high-penetration fully-connected transmission and distribution (T&D) co-simulation platform that supports the simulation of 10,000+ dispersed IBR models. The interfacing and iterative initialization techniques for the co-simulation have been implemented to maintain stable operation and simulation of large-multitude of IBR models. The phasor-domain IBR models with grid-forming (GFM) and grid-following (GFL) control are implemented in the distribution systems simulators. The developed platform is tested on high-performance computing (HPC) resources and can be utilized to explore the hierarchical control strategies of IBRs for the large-scale T&D hybrid system.
The Reactor Cavity Cooling System (RCCS) is an important passive safety system being incorporated into the overall safety strategy for high temperature advanced reactor concepts such as the High Temperature Gas- Cooled Reactors (HTGR). The Natural Convection Shutdown Heat Removal Test Facility (NSTF) at Argonne National Laboratory (Argonne) reflects a 1/2-scale model of the primary features of one conceptual air-cooled RCCS design. The project conducts ex-vessel, passive heat removal experiments in support of Department of Energy Office of Nuclear Energy’s Advanced Reactor Technology (ART) program, while also generating data for code validation purposes. While experiments are being conducted at the NSTF to evaluate the feasibility of the passive RCCS, parallel modeling and simulation efforts are ongoing to support the design, fabrication, and operation of these natural convection systems. Both system-level and high fidelity computational fluid dynamics (CFD) analyses were performed to gain a complete understanding of the complex flow and heat transfer phenomena in natural convection systems. This paper provides a summary of the RELAP5-3D NSTF model development efforts and provides comparisons between simulation results and experimental data from the NSTF. Overall, the simulation results compared favorably to the experimental data, however, further analyses need to be conducted to investigate any identified differences.
The Computational Structural Mechanics (CSM) activity is developing advanced structural analysis and computational methods that exploit high-performance computers. Methods are developed in the framework of the CSM testbed software system and applied to representative complex structural analysis problems from the aerospace industry. An overview of the CSM testbed methods development environment is presented and some numerical methods developed on a CRAY-2 are described. Selected application studies performed on the NAS CRAY-2 are also summarized.
The goal of this project was to develop a suite of molecular models to investigate adhesive performance as a result of the presence of different levels of moisture. These different levels of moisture are accepted in the adhesives community to represent different states of aging. First, our team modeled the adhesive’s crosslinked structure. Next, we modeled adding water to the crosslinked structure and used high performance computing resources to simulate water diffusion through the bulk adhesive and possible sites for chain scission. Finally, we modeled the response of the simulated adhesive to tensile loads at various levels of moisture content. These simulations were validated by experimental characterization by the industry partner. We found that the Young’s modulus of the adhesive decreased by about 30% when the water content increased to 3 wt.%. Our simulations also identified the likely site where the polymer chain could break. Our integration of molecular level simulations and mechanical property predictions using high performance computing with experimental studies of a hydrated epoxy system has yielded fundamental insights that will help industry advance lightweight joining technologies.
Understanding performance and provenance of task-based workflows poses significant challenges, particularly in distributed configurations where resources are shared by multiple applications. Task-based workflow management systems further complicate performance predictability because of their dynamicity that subtly alters task execution order from run to run. In this paper we propose a layered characterization framework for performance and task provenance for Dask.distributed workflows running on high-performance computing (HPC) platforms. It collects data from jobs, the workflow management system, and the operating system to aid in understanding the performance of these workflows. Our approach encompasses three main contributions: first, an extension of Dask.distributed to capture high-fidelity task provenance using Mochi data services; second, the adaptation of the established HPC I/O characterization tool Darshan to gather high-fidelity I/O data, thereby enhancing the granularity of our analysis; and third, a framework to combine and process the collected data and provide helpful insights into performance characterization and reproducibility, alongside our lessons learned.
The growing gap between sustained and peak performance for scientific applications has become a well-known problem in high performance computing. The recent development of parallel vector systems offers the potential to bridge this gap for a significant number of computational science codes and deliver a substantial increase in computing capabilities. This paper examines the intranode performance of the NEC SX6 vector processor and the cache-based IBM Power3/4 superscalar architectures across a number of key scientific computing areas. First, we present the performance of a microbenchmark suite that examines a full spectrum of low-level machine characteristics. Next, we study the behavior of the NAS Parallel Benchmarks using some simple optimizations. Finally, we evaluate the perfor- mance of several numerical codes from key scientific computing domains. Overall results demonstrate that the SX6 achieves high performance on a large fraction of our application suite and in many cases significantly outperforms the RISC-based architectures. However, certain classes of applications are not easily amenable to vectorization and would likely require extensive reengineering of both algorithm and implementation to utilize the SX6 effectively.
The Simulation EnviRonment for Geomorphology, Hydrodynamics, and Ecohydrology in Integrated form (SERGHEI) is a multi-dimensional, multi-domain, and multi-physics model framework for environmental and landscape simulation, designed with an outlook towards Earth system modelling. At the core of SERGHEI's innovation is its performance-portable high-performance parallel-computing (HPC) implementation, built from scratch on the Kokkos portability layer, allowing SERGHEI to be deployed, in a performance-portable fashion, in graphics processing unit (GPU)-based heterogeneous systems. In this work, we explore combinations of MPI and Kokkos using OpenMP and CUDA backends. In this contribution, we introduce the SERGHEI model framework and present with detail its first operational module for solving shallow-water equations (SERGHEI-SWE) and its HPC implementation. This module is designed to be applicable to hydrological and environmental problems including flooding and runoff generation, with an outlook towards Earth system modelling. Its applicability is demonstrated by testing several well-known benchmarks and large-scale problems, for which SERGHEI-SWE achieves excellent results for the different types of shallow-water problems. Finally, SERGHEI-SWE scalability and performance portability is demonstrated and evaluated on several TOP500 HPC systems, with very good scaling in the range of over 20 000 CPUs and up to 256 state-of-the art GPUs.
Explore the source record for details and available documents.