Search NASA⌕ Search

Engineering topics

de Supinski, Bronis R.

Publications and source records attributed to de Supinski, Bronis R..

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Performance on HPC Platforms Is Possible Without C++

Computing at large scales has become extremely challenging due to increasing heterogeneity in both hardware and software. More and more scientific workflows must tackle a range of scales and use machine learning and AI intertwined with more traditional numerical modeling methods, placing more demands on computational platforms. These constraints indicate a need to fundamentally rethink the way computational science is done and the tools that are needed to enable these complex workflows. The current set of C++-based solutions may not suffice, and relying exclusively upon C++ may not be the best option, especially because several newer languages and boutique solutions offer more robust design features to tackle the challenges of heterogeneity. In June 2023, we held a mini symposium that explored the use of newer languages and heterogeneity solutions that are not tied to C++ and that offer options beyond template metaprogramming and Parallel. For for performance and portability. In conclusion, we describe some of the presentations and discussion from the mini symposium in this article.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Non-Proprietary Companion to the Q1 CY2021 PathForward Final Assessment (WBS 2.4.1, Milestone PM-HI-1040)

The PathForward element of the Exascale Computing Project (ECP) Hardware and Integration (HI) focus area has met its goals of preparing the US industry for exascale system procurements and generally improving US competitiveness in the worldwide computing market. This report is a companion to the deliverable for the ECP milestone PM-HI-1040, Assess PathForward Impact Against Exascale Hardware Challenges, which gives a summary of the final status of each PathForward project, describes progress achieved against PathForward contract milestones, and includes a final assessment of each vendor’s progress on key exascale challenges. This companion document details the results of the PathForward research to the extent possible without disclosing proprietary information and the impact on products and US exascale systems. In addition, it captures lessons learned in order to inform future projects in general and in high-performance computing in particular.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Statistical and machine learning models for optimizing energy in parallel applications

Rising power costs and constraints are driving a growing focus on the energy efficiency of high performance computing systems. The unique characteristics of a particular system and workload and their effect on performance and energy efficiency are typically difficult for application users to assess and to control. Settings for optimum performance and energy efficiency can also diverge, so we need to identify trade-off options that guide a suitable balance between energy use and performance. We present statistical and machine learning models that only require a small number of runs to make accurate Pareto-optimal trade-off predictions using parameters that users can control. We study model training and validation using several parallel kernels and more complex workloads, including Algebraic Multigrid (AMG), Large-scale Atomic Molecular Massively Parallel Simulator, and Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Here, we demonstrate that we can train the models using as few as 12 runs, with prediction error of less than 10%. Our AMG results identify trade-off options that provide up to 45% improvement in energy efficiency for around 10% performance loss. We reduce the sample measurement time required for AMG by 90%, from 13 h to 74 min.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

The Design, Deployment, and Evaluation of the CORAL Pre-Exascale Systems

CORAL, the Collaboration of Oak Ridge, Argonne and Livermore, is fielding two similar IBM systems, Summit and Sierra, with NVIDIA GPUs that will replace the existing Titan and Sequoia systems. Summit and Sierra are currently ranked No. 1 and No. 3, respectively on the Top500 list. We discuss the design and key differences of the systems. Our evaluation of the systems highlights the following. Applications that fit in HBM see the most benefit and may prefer more GPUs; however, for some applications, the CPU-GPU bandwidth is more important than the number of GPUs. The node-local burst buffer scales linearly, and can achieve a 4X improvement over the parallel file system for large jobs; smaller jobs, however, may benefit from writing directly to the PFS. Finally, several CPU, network and memory bound analytics and GPU-bound deep learning codes achieve up to a 11X and 79X speedup/node, respectively over Titan.

Vazhkudai, Sudharshan↗
DOE OSTI2018

The Ongoing Evolution of OpenMP

Here, this paper presents an overview of the past, present and future of the OpenMP application programming interface (API). While the API originally specified a small set of directives that guided shared memory fork-join parallelization of loops and program sections, OpenMP now provides a richer set of directives that capture a wide range of parallelization strategies that are not strictly limited to shared memory. As we look toward the future of OpenMP, we immediately see further evolution of the support for that range of parallelization strategies and the addition of direct support for debugging and performance analysis tools. Looking beyond the next major release of the specification of the OpenMP API, we expect the specification eventually to include support for more parallelization strategies and to embrace closer integration into its Fortran, C and, in particular, C++ base languages, which will likely require the API to adopt additional programming abstractions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Application Modernization at LLNL and the Sierra Center of Excellence

We repport that in 2014, Lawrence Livermore National Laboratory began acquisition of Sierra, a pre-exascale system from IBM and Nvidia. It marks a significant shift in direction for LLNL by introducing the concept of heterogeneous computing via GPUs. LLNL’s mission requires application teams to prepare for this paradigm shift. Thus, the Sierra procurement required a proposed Center of Excellence that would align the expertise of the chosen vendors with laboratory personnel that represent the application developers, system software, and tool providers in a concentrated effort to prepare the laboratory’s codes in advance of the system transitioning to production in 2018. Finally, this article presents LLNL’s overall application strategy, with a focus on how LLNL is collaborating with IBM and Nvidia to ensure a successful transition of its mission-oriented applications into the exascale era.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

A survey on software methods to improve the energy efficiency of parallel computing

Energy consumption is one of the top challenges for achieving the next generation of supercomputing. Codesign of hardware and software is critical for improving energy efficiency (EE) for future large-scale systems. Many architectural power-saving techniques have been developed, and most hardware components are approaching physical limits. Accordingly, parallel computing software, including both applications and systems, should exploit power-saving hardware innovations and manage efficient energy use. In addition, new power-aware parallel computing methods are essential to decrease energy usage further. This article surveys software-based methods that aim to improve EE for parallel computing. It reviews the methods that exploit the characteristics of parallel scientific applications, including load imbalance and mixed precision of floating-point (FP) calculations, to improve EE. In addition, this article summarizes widely used methods to improve power usage at different granularities, such as the whole system and per application. In particular, it describes the most important techniques to measure and to achieve energy-efficient usage of various parallel computing facilities, including processors, memories, and networks. Overall, this article reviews the state-of-the-art of energy-efficient methods for parallel computing to motivate researchers to achieve optimal parallel computing under a power budget constraint.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Evaluating and extending user-level fault tolerance in MPI applications

The user-level failure mitigation (ULFM) interface has been proposed to provide fault-tolerant semantics in the Message Passing Interface (MPI). Previous work presented performance evaluations of ULFM; yet questions related to its programability and applicability, especially to non-trivial, bulk synchronous applications, remain unanswered. In this article, we present our experiences on using ULFM in a case study with a large, highly scalable, bulk synchronous molecular dynamics application to shed light on the advantages and difficulties of this interface to program fault-tolerant MPI applications. We found that, although ULFM is suitable for master–worker applications, it provides few benefits for more common bulk synchronous MPI applications. Furthermore, to address these limitations, we introduce a new, simpler fault-tolerant interface for complex, bulk synchronous MPI programs with better applicability and support than ULFM for application-level recovery mechanisms, such as global rollback.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Exploiting Redundancy and Application Scalability for Cost-Effective, Time-Constrained Execution of HPC Applications on Amazon EC2

The use of clouds to execute high-performance computing (HPC) applications has greatly increased recently. Clouds provide several potential advantages over traditional supercomputers and in-house clusters. The most popular cloud is currently Amazon EC2, which provides fixed-cost and variable-cost, auction-based options. The auction market trades lower cost for potential interruptions that necessitate checkpointing; if the market price exceeds the bid price, a node is taken away from the user without warning. We explore techniques to maximize performance per dollar given a time constraint within which an application must complete. Specifically, we design and implement multiple techniques to reduce expected cost by exploiting redundancy in the EC2 auction market. We then design an adaptive algorithm that selects a scheduling algorithm and determines the bid price. We show that our adaptive algorithm executes programs up to seven times cheaper than using the on-demand market and up to 44 percent cheaper than the best non-redundant, auction-market algorithm. We extend our adaptive algorithm to incorporate application scalability characteristics for further cost savings. In conclusion, we show that the adaptive algorithm informed with scalability characteristics of applications achieves up to 56 percent cost savings compared to the expected cost for the base adaptive algorithm run at a fixed, user-defined scale.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Diagnosis of Performance Faults in LargeScale MPI Applications via Probabilistic Progress-Dependence Inference

Debugging large-scale parallel applications is challenging. Most existing techniques provide little information about failure root causes. Further, most debuggers significantly slow down program execution, and run sluggishly with massively parallel applications. This paper presents a novel technique that scalably infers the tasks in a parallel program on which a failure occurred, as well as the code in which it originated. Our technique combines scalable runtime analysis with static analysis to determine the least-progressed task(s) and to identify the code lines at which the failure arose. We present a novel algorithm that infers probabilistically progress dependence among MPI tasks using a globally constructed Markov model that represents tasks' control-flow behavior. In comparison to previous work, our algorithm infers more precisely the least-progressed task. Further, we combine this technique with static backward slicing analysis, further isolating the code responsible for the current state. A blind study demonstrates that our technique isolates the root cause of a concurrency bug in a molecular dynamics simulation, which only manifests itself at 7,996 tasks or more. We extensively evaluate fault coverage of our technique via fault injections in 10 HPC benchmarks and show that our analysis takes less than a few seconds on thousands of parallel tasks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

CoreTSAR: Core Task-Size Adapting Runtime

Heterogeneity continues to increase at all levels of computing, with the rise of accelerators such as GPUs, FPGAs, and other co-processors into everything from desktops to supercomputers. As a consequence, efficiently managing such disparate resources has become increasingly complex. CoreTSAR seeks to reduce this complexity by adaptively worksharing parallel-loop regions across compute resources without requiring any transformation of the code within the loop. Lastly, our results show performance improvements of up to three-fold over a current state-of-the-art heterogeneous task scheduler as well as linear performance scaling from a single GPU to four GPUs for many codes. In addition, CoreTSAR demonstrates a robust ability to adapt to both a variety of workloads and underlying system configurations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Detailed Modeling and Evaluation of a Scalable Multilevel Checkpointing System

High-performance computing (HPC) systems are growing more powerful by utilizing more components. As the system mean time before failure correspondingly drops, applications must checkpoint frequently to make progress. But, at scale, the cost of checkpointing becomes prohibitive. A solution to this problem is multilevel checkpointing, which employs multiple types of checkpoints in a single run. Moreover, lightweight checkpoints can handle the most common failure modes, while more expensive checkpoints can handle severe failures. We designed a multilevel checkpointing library, the Scalable Checkpoint/Restart (SCR) library, that writes lightweight checkpoints to node-local storage in addition to the parallel file system. We present probabilistic Markov models of SCR's performance. We show that on future large-scale systems, SCR can lead to a gain in machine efficiency of up to 35 percent, and reduce the load on the parallel file system by a factor of two. In addition, we predict that checkpoint scavenging, or only writing checkpoints to the parallel file system on application termination, can reduce the load on the parallel file system by 20 × on today's systems and still maintain high application efficiency.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

Strategies for Energy-Efficient Resource Management of Hybrid Programming Models

Many scientific applications are programmed using hybrid programming models that use both message passing and shared memory, due to the increasing prevalence of large-scale systems with multicore, multisocket nodes. Previous work has shown that energy efficiency can be improved using software-controlled execution schemes that consider both the programming model and the power-aware execution capabilities of the system. However, such approaches have focused on identifying optimal resource utilization for one programming model, either shared memory or message passing, in isolation. The potential solution space, thus the challenge, increases substantially when optimizing hybrid models since the possible resource configurations increase exponentially. Nonetheless, with the accelerating adoption of hybrid programming models, we increasingly need improved energy efficiency in hybrid parallel applications on large-scale systems. In this work, we present new software-controlled execution schemes that consider the effects of dynamic concurrency throttling (DCT) and dynamic voltage and frequency scaling (DVFS) in the context of hybrid programming models. Specifically, we present predictive models and novel algorithms based on statistical analysis that anticipate application power and time requirements under different concurrency and frequency configurations. Furthermore, we apply our models and methods to the NPB MZ benchmarks and selected applications from the ASC Sequoia codes. Overall, we achieve substantial energy savings (8.74 percent on average and up to 13.8 percent) with some performance gain (up to 7.5 percent) or negligible performance loss.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

MPI Runtime Error Detection with MUST: Advances in Deadlock Detection

The widely used Message Passing Interface (MPI) is complex and rich. As a result, application developers require automated tools to avoid and to detect MPI programming errors. We present the Marmot Umpire Scalable Tool (MUST) that detects such errors with significantly increased scalability. We present improvements to our graph-based deadlock detection approach for MPI, which cover future MPI extensions. Our enhancements also check complex MPI constructs that no previous graph-based detection approach handled correctly. Finally, we present optimizations for the processing of MPI operations that reduce runtime deadlock detection overheads. Existing approaches often require 𝒪( p ) analysis time per MPI operation, for p processes. We empirically observe that our improvements lead to sub-linear or better analysis time per operation for a wide range of real world applications.

Hilbrich, Tobias↗
DOE OSTI2013

McrEngine: A Scalable Checkpointing System Using Data-Aware Aggregation and Compression

High performance computing (HPC) systems use checkpoint-restart to tolerate failures. Typically, applications store their states in checkpoints on a parallel file system (PFS). As applications scale up, checkpoint-restart incurs high overheads due to contention for PFS resources. The high overheads force large-scale applications to reduce checkpoint frequency, which means more compute time is lost in the event of failure. We alleviate this problem through a scalable checkpoint-restart system, mcrEngine. McrEngine aggregates checkpoints from multiple application processes with knowledge of the data semantics available through widely-used I/O libraries, e.g., HDF5 and netCDF, and compresses them. Our novel scheme improves compressibility of checkpoints up to 115% over simple concatenation and compression. Our evaluation with large-scale application checkpoints show that mcrEngine reduces checkpointing overhead by up to 87% and restart overhead by up to 62% over a baseline with no aggregation or compression.

Islam, Tanzima Zerin↗