Search NASA⌕ Search

SEARCH · Search NASA

Results for “HPC system”

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 343 records · Page 19

LibPowerMon: A vendor-neutral system monitoring library

As power becomes an important resource to provision for modern HPC systems it becomes crucial to obtain deeper insights into application's power and thermal characteristics and to correlate application context with important system-level measurements. Existing profiling tools to monitor power and thermal measurements either operate at a granularity that is not fine enough to correlate with application-level events that describe application context or are not equipped to sample application-level events transparently. We introduce LibPowerMon, a lightweight, vendor-neutral user-level profiling framework to simultaneously sample user-specified application events and system-level metrics at up to 1 kHz sampling intervals. At the application level, LibPowerMon provides a source-level phase markup interface to capture application context. At runtime level, LibPowerMon records MPI and OpenMP events and can be extended to include other runtimes. At the system level, LibPowerMon samples power, thermal and relevant metrics, and provides an extensible interface to system's control capabilities. In this release, LibPowerMon supports monitoring and control functionalities on the Intel, IBM Power9 and Nvidia Volta architectures.

MARATHE, ANIRUDDHAP↗

The HARNESS Workbench: Unified and Adaptive Access to Diverse HPC Platforms

The primary goal of the Harness WorkBench (HWB) project is to investigate innovative software environments that will help enhance the overall productivity of applications science on diverse HPC platforms. Two complementary frameworks were designed: one, a virtualized command toolkit for application building, deployment, and execution, that provides a common view across diverse HPC systems, in particular the DOE leadership computing platforms (Cray, IBM, SGI, and clusters); and two, a unified runtime environment that consolidates access to runtime services via an adaptive framework for execution-time and post processing activities. A prototype of the first was developed based on the concept of a 'system-call virtual machine' (SCVM), to enhance portability of the HPC application deployment process across heterogeneous high-end machines. The SCVM approach to portable builds is based on the insertion of toolkit-interpretable directives into original application build scripts. Modifications resulting from these directives preserve the semantics of the original build instruction flow. The execution of the build script is controlled by our toolkit that intercepts build script commands in a manner transparent to the end-user. We have applied this approach to a scientific production code (Gamess-US) on the Cray-XT5 machine. The second facet, termed Unibus, aims to facilitate provisioning and aggregation of multifaceted resources from resource providers and end-users perspectives. To achieve that, Unibus proposes a Capability Model and mediators (resource drivers) to virtualize access to diverse resources, and soft and successive conditioning to enable automatic and user-transparent resource provisioning. A proof of concept implementation has demonstrated the viability of this approach on high end machines, grid systems and computing clouds.

97 MATHEMATICS AND COMPUTING↗

Towards Acceptance Testing at the Exascale Frontier

At the 2007 Cray User Group meeting, the Oak Ridge Leadership Computing Facility (OLCF) introduced the OLCF Test Harness (OTH), a framework[1] used for acceptance testing of the Jaguar supercomputer[2]. Since then, the OTH framework has evolved to version 2.0 which adds new features and streamlines usability. The OTH is the key piece of software used to orchestrate acceptance testing for all OLCF computational resources before they are deployed for production use, including our leadership class high performance computing (HPC) systems. The OTH framework is written in Python and is publicly available[3].In this paper, we first describe the requirements, design, and structure of the OTH. Then, we present specific improvements developed to support acceptance testing of the OLCF’s Summit system[4]. We will also showcase new OTH features that have been added to streamline the acceptance test process as well as the motivation behind those changes. As part of this work, we also evaluated different workflow tools in order to determine whether these tools could complement the OTH in two key areas: automation and reporting. The advantages and disadvantages identified with each tool will be discussed. Lastly, we summarize the challenges and lessons learned collected from using the OTH for the acceptance of the last three flagship systems at the OLCF. These may be useful for other HPC centers developing their own testing frameworks or those interested in using the OTH.

Melesse Vergara, Veronica↗

A Quantitative Approach to Architecting All-Flash Lustre File Systems

New experimental and AI-driven workloads are moving into the realm of extreme-scale HPC systems at the same time that high-performance flash is becoming cost-effective to deploy at scale. This confluence poses a number of new technical and economic challenges and opportunities in designing the next generation of HPC storage and I/O subsystems to achieve the right balance of bandwidth, latency, endurance, and cost. In this work, we present quantitative models that use workload data from existing, disk-based file systems to project the architectural requirements of all-flash Lustre file systems. Using data from NERSC’s Cori I/O subsystem, we then demonstrate the minimum required capacity for data, capacity for metadata and data-on-MDT, and SSD endurance for a future all-flash Lustre file system.

Lockwood, GK↗

The Case for A Hierarchal System Model for Linux Clusters

The computer industry today is no longer driven, as it was in the 40s, 50s and 60s, by High-performance computing requirements. Rather, HPC systems, especially Leadership class systems, sit on top of a pyramid investment mode. Figure 1 shows a representative pyramid investment model for systems hardware. At the base of the pyramid is the huge investment (order 10s of Billions of US Dollars per year) in semiconductor fabrication and process technologies. These costs, which are approximately doubling with every generation, are funded from investments multiple markets: enterprise, desktops, games, embedded and specialized devices. Over and above these base technology investments are investments for critical technology elements such as microprocessor, chipsets and memory ASIC components. Investments for these components are spread across the same markets as the base semiconductor processes investments. These second tier investments are approximately half the size of the lower level of the pyramid. The next technology investment layer up, tier 3, is more focused on scalable computing systems such as those needed for HPC and other markets. These tier 3 technology elements include networking (SAN, WAN and LAN), interconnects and large scalable SMP designs. Above these is tier 4 are relatively small investments necessary to build very large, scalable systems high-end or Leadership class systems. Primary among these are the specialized network designs of vertically integrated systems, etc.

99 GENERAL AND MISCELLANEOUS↗

A Brief Survey on High Performance Computing Systems Power Management

This paper provides a survey of software-based power management techniques in High Performance Computing (HPC) systems. Seven existing power management and monitoring tools and frameworks are discussed. These are: Variorum, dynamic energy-performance optimizer (DEPO), Powersched, Bull Dynamic Power Optimizer (BDPO), Energy Aware Runtime (EAR), Global Extensible Open Power Manager (GEOPM), and PoLiMEr. Each of these tools is evaluated based on hardware abstraction, optimization methods, usability, and experimental validation. This survey highlights the diversity of approaches in managing energy efficiency, from vendor-neutral APIs to algorithm-driven power capping, and dynamic frequency adjustments. Given that energy requirements for large computational systems is increasing quickly, the importance of integrating these tools into existing HPC environments and the need for further research in this rapidly evolving field is also discussed.

97 - MATHEMATICS AND COMPUTING↗

Data Jockey: Automatic Data Management for HPC Multi-Tiered Storage Systems

We present the design and implementation of Data Jockey, a data management system for HPC multi-tiered storage systems. As a centralized data management control plane, Data Jockey automates bulk data movement and placement for scientific workflows and integrates into existing HPC storage infrastructures. Data Jockey simplifies data management by eliminating human effort in programming complex data movements, laying datasets across multiple storage tiers when supporting complex workflows, which in turn increases the usability of multitiered storage systems emerging in modern HPC data centers.Specifically, Data Jockey presents a new data management scheme called “goal driven data management” that can automatically infer low-level bulk data movement plans from declarative high-level goal statements that come from the lifetime of iterative runs of scientific workflows. While doing so, Data Jockey aims to minimize data wait times by taking responsibility for datasets that are unused or to be used, and aggressively utilizing the capacity of the upper, higher performant storage tiers.We evaluated a prototype implementation of Data Jockey under a synthetic workload based on a year’s worth of Oak Ridge Leadership Computing Facility’s (OLCF) operational logs. Our evaluations suggest that Data Jockey leads to higher utilization of the upper storage tiers while minimizing the programming effort of data movement compared to human involved, per-domain adhoc data management scripts.

Shin, Woong↗

Performance and energy impact of OpenMP runtime configurations on power constrained systems

Energy consumption of an application has attained the same vitality as execution time in current HPC systems, primarily due to the massive energy consumption of these systems and consequently substantial energy bills. This fact forced software developers as well as hardware vendors to put an extensive focus on energy consumption. This focus becomes even more crucial in power constrained systems, where the amount of power dedicated to applications is limited. In such systems, finding the optimal balance between execution time and energy consumption has a significant impact on utilizing these systems efficiently. A programming model is a necessary layer where optimizations can be applied to find the optimal balance. OpenMP is a standard programming model for intra-node parallelism. In this paper, we give a detailed analysis of how to optimize the execution time and energy consumption of an OpenMP application using the OpenMP runtime environment at different power levels. We studied twenty-eight OpenMP parallel regions from five NAS Parallel Benchmark (NPB) applications. Based on the study, we show that for a given power level, a suitable selection of OpenMP runtime parameters can improve the execution time and energy consumption of a parallel region up to 67% and 72%, respectively. We also show that these fine-grained improvements resulted in up to 26% of execution time and 38% of energy consumption improvement for a given OpenMP application. We finally observed that optimizing the energy consumption behavior alone across all power levels regardless of execution time behavior can lead to an additional energy consumption improvement of up to 55%. However, this improvement comes at the expense of execution time.

97 MATHEMATICS AND COMPUTING↗

Exploiting communication concurrency on high performance computing systems

Although logically available, applications may not exploit enough instantaneous communication concurrency to maximize hardware utilization on HPC systems. This is exacerbated in hybrid programming models such as SPMD+OpenMP. We present the design of a "multi-threaded" runtime able to transparently increase the instantaneous network concurrency and to provide near saturation bandwidth, independent of the application configuration and dynamic behavior. The runtime forwards communication requests from application level tasks to multiple communication servers. Our techniques alleviate the need for spatial and temporal application level message concurrency optimizations. Experimental results show improved message throughput and bandwidth by as much as 150% for 4KB bytes messages on InfiniBand and by as much as 120% for 4KB byte messages on Cray Aries. For more complex operations such as all-to-all collectives, we observe as much as 30% speedup. This translates into 23% speedup on 12,288 cores for a NAS FT implemented using FFTW. We also observe as much as 76% speedup on 1,500 cores for an already optimized UPC+OpenMP geometric multigrid application using hybrid parallelism.

97 MATHEMATICS AND COMPUTING↗

Confidence: Analyzing Performance With Empirical Probabilities

Variability in the performance of shared system components is a major obstacle in analyzing the effective throughput of leadership class computers. Shared file systems and networks are serious impediments to achieving repeatable application performance on HPC systems. In particular, performance analysts are likely to be interested in quantifying differences between average-case behavior, worst-case behavior, and standard deviation for shared system components. Typical descriptions of these statistics assume a normal distribution; however, in non-linear and multi-modal performance distributions, summary statistics are often misleading. In this paper we describe Confidence, a tool for analyzing the full spectrum of performance for a benchmarking code. By including all of the experimental outcomes in the analysis without discarding any measurements, Confidence enables a novel analysis of benchmark performance. Finally, we use Confidence to characterize the network performance of three high performance computer systems, and we provide recommendations for leveraging network performance in latency sensitive applications.

Settlemyer, Bradley W↗

Distributing Power Grid State Estimation on HPC Clusters A System Architecture Prototype

The future power grid is expected to further expand with highly distributed energy sources and smart loads. The increased size and complexity lead to increased burden on existing computational resources in energy control centers. Thus the need to perform real-time assessment on such systems entails efficient means to distribute centralized functions such as state estimation in the power system. In this paper, we present our early prototype of a system architecture that connects distributed state estimators individually running parallel programs to solve non-linear estimation procedure. The prototype consists of a middleware and data processing toolkits that allows data exchange in the distributed state estimation. We build a test case based on the IEEE 118 bus system and partition the state estimation of the whole system model to available HPC clusters. The measurement from the testbed demonstrates the low overhead of our solution.

Distributed architecture↗

Building 725 Expansion

In October 2017, Sandia broke ground for a new computing center dedicated to High Performance Computing. The east expansion of Building 725 was entirely conceived of, designed, and built in less than 18 months and is a certified LEED Gold design building, the first of its kind for a data center in the State of New Mexico. This 15,000 square-foot building, with novel energy and water-saving technologies, will house Astra, the first in a new generation of Advanced Architecture Prototype Systems to be deployed by the NNSA and the first of many HPC systems in Building 725 East.

99 GENERAL AND MISCELLANEOUS↗

Toward designing effective exascale scientific computing workflows: experiences and best practices

Many fields within scientific computing have embraced advances in big-data analysis and machine learning, which often requires the deployment of large, distributed and complicated workflows that may combine training neural networks, performing simulations, running inference, and performing database queries and data analysis in asynchronous, parallel and pipelined execution frameworks. Such a shift has brought into focus the need for scalable, efficient workflow management solutions with reproducibility, error and provenance handling, traceability, and checkpoint-restart capabilities, among other needs. Here, we discuss challenges and best-practices for deploying exascale-generation computational science workflows on resources at the Oak Ridge Leadership Computing Facility (OLCF). We present our experiences with large-scale deployment of distributed workflows on the Summit supercomputer, including for bioinformatics and computational biophysics, materials science, and deep learning model optimization. We also present problems and solutions created by working within a Python-centric software base on traditional HPC systems, and discuss steps that will be required before the convergence of HPC, AI, and data science can be fully realized. Our results point to a wealth of exciting new possibilities for harnessing this convergence to tackle new scientific challenges.

Coletti, Mark↗

iez: Resource Contention Aware Load Balancing for Large-Scale Parallel File Systems

Parallel I/O performance is crucial to sustaining scientific applications on large-scale High-Performance Computing (HPC) systems. However, I/O load imbalance in the underlying distributed and shared storage systems can significantly reduce overall application performance. There are two conflicting challenges to mitigate this load imbalance: (i) optimizing systemwide data placement to maximize the bandwidth advantages of distributed storage servers, i.e., allocating I/O resources efficiently across applications and job runs; and (ii) optimizing client-centric data movement to minimize I/O load request latency between clients and servers, i.e., allocating I/O resources efficiently in service to a single application and job run. Moreover, existing approaches that require application changes limit wide-spread adoption in commercial or proprietary deployments. We propose iez, an “end-to-end control plane” where clients transparently and adaptively write to a set of selected I/O servers to achieve balanced data placement. Our control plane leverages realtime load information for distributed storage server global data placement while our design model leverages trace-based optimization techniques to minimize I/O load request latency between clients and servers. We evaluate our proposed system on an experimental cluster for two common use cases: synthetic I/O benchmark IOR for large sequential writes and a scientific application I/O kernel, HACC-I/O. Results show read and write performance improvements of up to 34% and 32%, respectively, compared to the state of the art.

Wadhwa, Bharti↗

2009 fault tolerance for extreme-scale computing workshop, Albuquerque, NM - March 19-20, 2009.

This is a report on the third in a series of petascale workshops co-sponsored by Blue Waters and TeraGrid to address challenges and opportunities for making effective use of emerging extreme-scale computing. This workshop was held to discuss fault tolerance on large systems for running large, possibly long-running applications. The main point of the workshop was to have systems people, middleware people (including fault-tolerance experts), and applications people talk about the issues and figure out what needs to be done, mostly at the middleware and application levels, to run such applications on the emerging petascale systems, without having faults cause large numbers of application failures. The workshop found that there is considerable interest in fault tolerance, resilience, and reliability of high-performance computing (HPC) systems in general, at all levels of HPC. The only way to recover from faults is through the use of some redundancy, either in space or in time. Redundancy in time, in the form of writing checkpoints to disk and restarting at the most recent checkpoint after a fault that cause an application to crash/halt, is the most common tool used in applications today, but there are questions about how long this can continue to be a good solution as systems and memories grow faster than I/O bandwidth to disk. There is interest in both modifications to this, such as checkpoints to memory, partial checkpoints, and message logging, and alternative ideas, such as in-memory recovery using residues. We believe that systematic exploration of these ideas holds the most promise for the scientific applications community. Fault tolerance has been an issue of discussion in the HPC community for at least the past 10 years; but much like other issues, the community has managed to put off addressing it during this period. There is a growing recognition that as systems continue to grow to petascale and beyond, the field is approaching the point where we don't have any choice but to address this through R&D efforts.

97 MATHEMATICS AND COMPUTING↗

Dynamic Resource Management for Parallel Tasks in an Oversubscribed Energy-Constrained Heterogeneous Environment

Abstract: The worth of completing parallel tasks is modeled using utility functions, which monotonically-decrease with time and represent the importance and urgency of a task. These functions define the utility earned by a task at the time of its completion. The performance of such a system is measured as the total utility earned by all completed tasks over some interval of time (e.g., 24 hours). To maximize system performance when scheduling dynamically arriving parallel tasks onto a high performance computing (HPC) system that is oversubscribed and energy-constrained, we have designed, analyzed, and compared different heuristic techniques. Four utility-aware heuristics (i.e., Max Utility, Max Utility-per-Time, Max Utility-per-Resource, and Max Utility-per-Energy), three FCFS-based heuristics (Conservative Backfilling, EASY Backfilling, and FCFS with Multiple Queues), and a Random heuristic were examined in this study. A technique that is often used with the FCFS-based heuristics is the concept of a permanent reservation. We compare the performance of permanent reservations with temporary place-holders to demonstrate the advantages that place-holders can provide. We also present a novel energy filtering technique that constrains the maximum energy-per-resource used by each task. We conducted a simulation study to evaluate the performance of these heuristics and techniques in an energy-constrained oversubscribed HPC environment. With place-holders, energy filtering, and dropping tasks with low potential utility, our utility-aware heuristics are able to significantly outperform the existing FCFS-based techniques.

Energy-Constrained↗

Darshan for HEP applications

Modern HEP workflows must manage increasingly large and complex data collections. HPC facilities may be employed to help meet these workflows’ growing data processing needs. However, a better understanding of the I/O patterns and underlying bottlenecks of these workflows is necessary to meet the performance expectations of HPC systems.Darshan is a lightweight I/O characterization tool that captures concise views of HPC application I/O behavior. It intercepts application I/O calls at runtime, records file access statistics for each process, and generates log files detailing application I/O access patterns.Typical HEP workflows include event generation, detector simulation, event reconstruction, and subsequent analysis stages. A study of the I/O behavior of the ATLAS simulation and filtering stage, and the CMS simulation workflow using Darshan is presented, including insights into the I/O operations and data access size.

Wang, Rui↗

Wattchmen: Watching the Wattchers – High Fidelity, Flexible GPU Energy Modeling

Modern GPU-rich HPC systems are increasingly becoming energy-constrained. Thus, understanding an application’s energy consumption becomes essential. Unfortunately, current GPU energy attribution techniques are either inaccurate, inflexible, or outdated. Therefore, we propose Wattchmen, a flexible methodology for measuring, attributing, and predicting GPU energy consumption. We construct a per-instruction energy model using a diverse set of microbenchmarks to systematically quantify the energy consumption of GPU instructions, enabling finer-grain prediction and energy consumption breakdowns for applications. Compared with the state-of-the-art systems like AccelWattch (32%) and Guser (25%), across 16 popular GPGPU, graph analytics, HPC, and ML workloads, Wattchmen reduces the mean absolute percent error (MAPE) to 14% on V100 GPUs. Furthermore, we show that Wattchmen provides similar MAPEs for water-cooled V100s (15%) and extends to later architectures, including air-cooled A100 (11%) and H100 (12%) GPUs. Finally, to further demonstrate Wattchmen ’s value, we apply it to applications such as Backprop and QMCPACK, where Wattchmen ’s insights enable energy reductions of up to 35%.

Tran, Brandon [University of Wisconsin, Madison] (↗