Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “workload”

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 613 records · Page 34

Fast Fault Injection and Sensitivity Analysis for Collective Communications, In: 2015 IEEE International Conference on Cluster Computing

The collective communication operations, which are widely used in parallel applications for global communication and synchronization are critical for application's performance and scalability. However, how faulty collective communications impact the application and how errors propagate between the application processes is largely unexplored. One of the critical reasons for this situation is the lack of fast evaluation method to investigate the impacts of faulty collective operations. The traditional random fault injection methods relying on a large amount of fault injection tests to ensure statistical significance require a significant amount of resources and time. These methods result in prohibitive evaluation cost when applied to the collectives. In this paper, we introduce a novel tool named Fast Fault Injection and Sensitivity Analysis Tool (FastFIT) to conduct fast fault injection and characterize the application sensitivity to faulty collectives. The tool achieves fast exploration by reducing the exploration space and predicting the application sensitivity using Machine Learning (ML) techniques. A basis for these techniques are implicit correlations between MPI semantics, application context, critical application features, and application responses to faulty collective communications. The experimental results show that our approach reduces the fault injection points and tests by 97% for representative benchmarks (NAS Parallel Benchmarks (NPB)) and a realistic application (Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS)) on a production supercomputer. Further, we statistically generalize the application sensitivity to faulty collective communications for these workloads, and present correlation between application features and the sensitivity.

Computer Science↗

A Scalable Distributed Louvain Algorithm for Large-Scale Graph Community Detection

We present a new distributed community detection algorithm for large graphs based on the Louvain method. We exploit a distributed delegate partitioning to ensure the workload and communication balancing among processors. In addition, we design a new heuristic strategy to carefully coordinate the community constitution in a distributed environment, and ensure the convergence of the distributed clustering algorithm. Our intensive experimental study has demonstrated the scalability and the correctness of our algorithm with various large-scale real-world and synthetic graph datasets using up to 32,768 processors.

Computer Science↗

Hierarchical Clock Synchronization in MPI

We present a new distributed community detection algorithm for large graphs based on the Louvain method. We exploit a distributed delegate partitioning to ensure the workload and communication balancing among processors. In addition, we design a new heuristic strategy to carefully coordinate the community constitution in a distributed environment, and ensure the convergence of the distributed clustering algorithm. Our intensive experimental study has demonstrated the scalability and the correctness of our algorithm with various large-scale real-world and synthetic graph datasets using up to 32,768 processors

Computer Science↗

A Quantitative Study of Deep Learning Training on Heterogeneous Supercomputers

Deep learning (DL) has become a key technique for solving complex problems in scientific research and discovery. DL training for science is substantially challenging because it has to deal with massive quantities of multi-dimensional data. High- performance computing (HPC) supercomputers are increasingly being employed for meeting the exponentially growing demand for DL. Multiple GPUs and high-speed interconnect network are needed for supporting DL on HPC systems. However, the excessive use of GPUs without considering effective benefits leads to inefficient resource utilization of these expensive setups. In this paper, we conduct a quantitative analysis to gauge the efficacy of DL workloads on the latest HPC system and identify viability of next-generation DL-optimized heterogeneous supercomputers for enabling researchers to develop more efficient resource management and distributed DL middleware. We evaluate well- known DL models with large-scale datasets using the popular TensorFlow framework, and provide a thorough evaluation in- cluding scalability, accuracy, variability, storage resource, GPU- GPU/GPU-CPU data transfer, and GPU utilization. Our analysis reveals that the latest heterogeneous supercomputing cluster shows varying performance trend as compared to the existing literature for single- and multi-node training. To the best of our knowledge, this is the first work to conduct such a quantitative and comprehensive study of DL training on a supercomputing system with multiple GPUs.

Han, Jingoo↗

ProvLight: Efficient Workflow Provenance Capture on the Edge-to-Cloud Continuum

Modern scientific workflows require hybrid infrastructures combining numerous decentralized resources on the IoT/Edge interconnected to Cloud/HPC systems (aka the Computing Continuum) to enable their optimized execution. Understanding and optimizing the performance of such complex Edge-to-Cloud workflows is challenging. Capturing the provenance of key performance indicators, with their related data and processes, may assist in understanding and optimizing workflow executions. However, the capture overhead can be prohibitive, particularly in resource-constrained devices, such as the ones on the IoT/Edge.To address this challenge, based on a performance analysis of existing systems, we propose ProvLight, a tool to enable efficient provenance capture on the IoT/Edge. We leverage simplified data models, data compression and grouping, and lightweight transmission protocols to reduce overheads. We further integrate ProvLight into the E2Clab framework to enable workflow provenance capture across the Edge-to-Cloud Continuum. This integration makes E2Clab a promising platform for the performance optimization of applications through reproducible experiments.We validate ProvLight at a large scale with synthetic workloads on 64 real-life IoT/Edge devices in the FIT IoT LAB testbed. Evaluations show that ProvLight outperforms state-of-the-art systems like ProvLake and DfAnalyzer in resource-constrained devices. ProvLight is 26—37x faster to capture and transmit provenance data; uses 5—7x less CPU; 2x less memory; transmits 2x less data; and consumes 2—2.5x less energy. ProvLight [1] and E2Clab [2] are available as open-source tools.

Rosendo, Daniel↗

Understanding Mixed Precision GEMM with MPGemmFI: Insights into Fault Resilience

Emerging deep learning workloads urgently need fast general matrix multiplication (GEMM). Thus, one of the critical features of machine-learning-specific accelerators such as NVIDIA Tensor Cores, AMD Matrix Cores, and Google TPUs is the support of mixed-precision enabled GEMM. For DNN models, lower-precision FP data formats and computation offer acceptable correctness but significant performance, area, and memory footprint improvement. While promising, the mixed-precision computation on error resilience remains unexplored. To this end, we develop a fault injection framework that systematically injects fault into the mixed-precision computation results. We investigate how the faults affect the accuracy of machine learning applications. Based on the characteristics of error resilience, we offer lightweight error detection and correction solutions that significantly improve the overall model accuracy by 75% if the models experience hardware faults. The solutions can be efficiently integrated into the accelerator's pipelines.

Fang, Bo↗

Machine Learning Models for GPU Error Prediction in a Large Scale HPC System

GPUs are widely deployed on large-scale HPC systems to provide powerful computational capability for scientific applications from various domains. As those applications are normally long-running, investigating the characteristics of GPU errors becomes imperative for reliability. In this paper, we first study the system conditions that trigger GPU errors using six-month trace data collected from a large-scale, operational HPC system. Then, we use machine learning to predict the occurrence of GPU errors, by taking advantage of temporal and spatial dependencies of the trace data. The resulting machine learning prediction framework is robust and accurate under different workloads.

Nie, Bin↗

Management, Analysis, and Visualization of Experimental and Observational Data – The Convergence of Data and Computing

Scientific user facilities—particle accelerators, telescopes, colliders, supercomputers, light sources, sequencing facilities, and more—operated by the U.S. Department of Energy (DOE) Office of Science (SC) generate ever increasing volumes of data at unprecedented rates from experiments, observations, and simulations. At the same time there is a growing community of experimentalists that require real-time data analysis feedback, to enable them to steer their complex experimental instruments to optimized scientific outcomes and new discoveries. Recent efforts in DOE-SC have focused on articulating the data-centric challenges and opportunities facing these science communities. Key challenges include difficulties coping with data size, rate, and complexity in the context of both real-time and post-experiment data analysis and interpretation. Solutions will require algorithmic and mathematical advances, as well as hardware and software infrastructures that adequately support data-intensive scientific workloads. This paper presents the summary findings of a workshop held by DOE-SC in September 2015, convened to identify the major challenges and the research that is needed to meet those challenges.

Bethel, E. Wes↗

High-Throughput Computing on High-Performance Platforms: A Case Study

The computing systems used by LHC experiments has historically consisted of the federation of hundreds to thousands of distributed resources, ranging from small to mid-size resource. In spite of the impressive scale of the existing distributed computing solutions, the federation of small to mid-size resources will be insufficient to meet projected future demands. This paper is a case study of how the ATLAS experiment has embraced Titan -- a DOE leadership facility in conjunction with traditional distributed high- throughput computing to reach sustained production scales of approximately 52M core-hours a years. The three main contributions of this paper are: (i) a critical evaluation of design and operational considerations to support the sustained, scalable and production usage of Titan; (ii) a preliminary characterization of a next generation executor for PanDA to support new workloads and advanced execution modes; and (iii) early lessons for how current and future experimental and observational systems can be integrated with production supercomputers and other platforms in a general and extensible manner.

Oleynik, D↗

Enabling Scientific Applications with Performance-Portability and High-Productivity for Multi-GPU Programming with JACC.Multi

This work bridges the gap between multi-GPU computing and high-productivity, performance-portable programming solutions. Our goal is to enhance scientific applications with a productive and portable solution—program once, deploy everywhere—for multi-GPU programming with no cost to programmability. To accomplish this, we implemented JACC.Multi, which is part of the Julia for ACCelerators (JACC) performance-portable framework. JACC. Multi is the only high-level, portable metaprogramming solution that targets multi-GPU environments and is integrated in a readily accessible programming language (e.g., Julia language). With transparent GPU-to-GPU communication, JACC. Multi is optimized for scientific application workloads and is portable for NVIDIA and AMD accelerators. For the evaluation, we use two modern multi-GPU systems: Hudson, which features two NVIDIA H100 Hopper GPUs per node, and Frontier, which features four AMD MI250X GPUs per node, each with two Graphics Compute Dies (GCDs) for a total of eight GCDs per node. Additionally, as part of the evaluation, we use JACC (one GPU), MPI+JACC, and JACC. Multi codes that implement well-known and widely used scientific algorithms/kernels such as the conjugate gradient algorithm and an explicit forward Euler solver that requires GPU-to-GPU communication. Overall, JACC. Multi codes achieve better performance than MPI+JACC codes and significant speedups over JACC (one GPU), with up to 1.9× on Hudson and 6× on Frontier.

Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗

Modular HPC I/O characterization with Darshan

Contemporary high-performance computing (HPC) applications encompass a broad range of distinct I/O strategies and are often executed on a number of different compute platforms in their lifetime. These large-scale HPC platforms employ increasingly complex I/O subsystems to provide a suitable level of I/O performance to applications. Tuning I/O workloads for such a system is nontrivial, and the results generally are not portable to other HPC systems. I/O profiling tools can help to address this challenge, but most existing tools only instrument specific components within the I/O subsystem that provide a limited perspective on I/O performance. The increasing diversity of scientific applications and computing platforms calls for greater flexibililty and scope in I/O characterization.

Snyder, Shane↗

Accelerating the Global Arrays ComEx Runtime using Multiple Progress Ranks

Partitioned Global Address Space (PGAS) models are part of system software that are being designed to support communication runtime for exascale applications. Suitability of using MPI as a scalable PGAS communication subsystem has been shown as a viable option due to its standardization and higher performance. We used MPI two-sided semantics with a combination of automatic and user defined splitting of MPI communicators to achieve asynchronous progress. Our implementation can make use of multiple asynchronous progress ranks (PR) per node that can be mapped to the computing architecture of a node in a distributed cluster. We are able to show significant speed up of over 2.0X and scaling of communication bound computational chemistry application over 1024 nodes of state- of-the-art HPC clusters. Our results show that while running a communication bound application workload on a certain number of cluster nodes, a use of optimum number of ranks dedicated for communication may be used to achieve asynchronous communication progress and obtain highest performance.

PGAS, global-arrays, NWCHEM, NWChemEx, HPC, GPGPU↗

Exploring Manycore Multinode Systems for Irregular Applications with FPGA Prototyping

We present a prototype of a multi-core architecture implemented on FPGA, designed to enable efficient execution of irregular applications on distributed shared memory machines, while maintaining high performance on regular workloads. The architecture is composed of off-the-shelf soft-core cores, local interconnection and memory interface, integrated with custom components that optimize it for irregular applications. It relies on three key elements: a global address space, multithreading, and fine-grained synchronization. Global addresses are scrambled to reduce the formation of network hot-spots, while the latency of the transactions is covered by integrating an hardware scheduler within the custom load/store buffers to take advantage from the availability of multiple executions threads, increasing the efficiency in a transparent way to the application. We evaluated a dual node system irregular kernels showing scalability in the number of cores and threads.

FPGA↗

Did the GPU obfuscate the load imbalance in my MPI simulation?

The current proliferation of GPU-based HPC systems necessitates a method for assessing the performance of simulations on heterogeneous machines. The addition of GPUs to a system adds multiple hierarchical levels of parallelism to the node architecture. In this paper, we demonstrate that the traditional load imbalance metric is insufficient for capturing the load imbalance on GPU-based machines, since it treats the GPU as a monolithic entity and ignores the internal parallelism. We propose a new hierarchical metric that improves the correlation of measured performance and application workload by up to 20.61%. Using our metric for determining application load instead of the traditional metric as the input for the load balancing algorithm reduces the residual load imbalance by up to 4× in our application.

Eberius, David↗

Integrating External Resources with a Task-Based Programming Model

Accessing external resources (e.g., loading input data, checkpointing snapshots, and out-of-core processing) can have a significant impact on the performance of supercomputer applications. However, no existing programming systems for high-performance computing directly manage and optimize these external accesses. As a result, users must explicitly manage external accesses alongside their computation at the application level, which can result in both correctness and performance issues. We address this limitation by introducing Iris, a task-based programming model with semantics for external resources. Iris allows applications to describe their access requirements to external resources and the relationship of those accesses to the computation. Iris incorporates external I/O into a deferred execution model, reschedules external I/O to overlap I/O with computation, and reduces external I/O when possible. We evaluate Iris on three microbenchmarks representative of important workloads in HPC and a full combustion simulation, S3D. We demonstrate that the Iris implementation of S3D reduces the external I/O overhead by up to 20×, compared to the Legion and the Fortran implementations.

97 MATHEMATICS AND COMPUTING↗

Scalable Proximity-Based Methods for Large-Scale Analysis of Atom Probe Data

Powered by recent advances in data acquisition technologies, today's state-of-the-art atom probe microscopes yield data sets with sizes ranging from a few million atoms to billions of atoms. Analysis of these atomic data sets within rea-sonable turnaround times is a pressing data analysis challenge for material scientists currently equipped with software systems that do not scale to these massive data sets. Here, we present the shared memory component of a larger ongoing effort to develop a multi-feature data analysis framework capable of analyzing atom probe data of all sizes and scales from desktop multicore machines to large-scale high-performance computing platforms with hybrid (shared and distributed memory) architectures. Our focus here is on a broad class of popular atom probe data analysis methods that rely on core time-consuming k-NN queries. We present a scalable, heuristic algorithm for k-NN queries using three-dimensional range trees. To demonstrate its efficacy, the k-NN algorithm is integrated with two use cases of atom probe data analysis methods and the resulting analysis times are shown to speedup by over 20X on a 32-core Cray XC40 node using workloads up to 8 million atoms, which is already beyond the at-scale capabilities of existing atom probe software. Using this k-NN algorithm, we also introduce a novel parameter estimation method for a class of cluster finding methods, called friends-of-friends (FoF) methods, to completely bypass their expensive pre-processing steps. In each case, we validate the results on a variety of control data sets.

Lu, Hao↗

A Parallel Alternative for Energy-Efficient Neural Network Training and Inferencing

Energy efficiency of training and inferencing with large neural network models is a critical challenge facing the future of sustainable large-scale machine learning workloads. This paper introduces an alternative strategy, called phantom parallelism, to minimize the net energy consumption of traditional tensor (model) parallelism, the most energy-inefficient component of large neural network training. The approach is presented in the context of feed-forward network architectures as a preliminary, but comprehensive, proof-of-principle study of the proposed methodology. We derive new forward and backward propagation operators for phantom parallelism, implement them as custom autograd operations within an end-to-end phantom parallel training pipeline and compare its parallel performance and energy-efficiency against those of conventional tensor parallel training pipelines. Formal analyses that predict lower bandwidth and FLOP counts are presented with supporting empirical results on up to 256 GPUs that corroborate these gains. Experiments are shown to deliver ∼50% reduction in the energy consumed to train FFNs using the proposed phantom parallel approach when compared with conventional tensor parallel methods. Additionally, the proposed approach is shown to train smaller phantom models to the same model loss on smaller GPU counts as larger tensor parallel models on larger GPU counts offering the possibility for even greater energy savings.

Seal, Sudip [ORNL] (ORCID:0000000332330656)↗

Efficient Extraction of High Centrality Vertices in Distributed Graphs

Betweenness centrality (BC) is an important measure for identifying high value or critical vertices in graphs, in variety of domains such as communication networks, road networks, and social graphs. However, calculating betweenness values is prohibitively expensive and, more often, domain experts are interested only in the vertices with the highest centrality values. In this paper, we first propose a partition-centric algorithm (MS-BC) to calculate BC for a large distributed graph that optimizes resource utilization and improves overall performance. Further, we extend the notion of approximate BC by pruning the graph and removing a subset of edges and vertices that contribute the least to the betweenness values of other vertices (MSL-BC), which further improves the runtime performance. We evaluate the proposed algorithms using a mix of real-world and synthetic graphs on an HPC cluster and analyze its strengths and weaknesses. The experimental results show an improvement in performance of upto 12x for large sparse graphs as compared to the state-of-the-art, and at the same time highlights the need for better partitioning methods to enable a balanced workload across partitions for unbalanced graphs such as small-world or power-law graphs.

97 MATHEMATICS AND COMPUTING↗