Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed memory”

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 163 records · Page 9

Exploring MPI Communication Models for Graph Applications Using Graph Matching as a Case Study

Traditional implementations of parallel graph operations on distributed memory platforms are written using Message Passing Interface (MPI) point to point communication primitives such as Send-Recv (blocking and nonblocking). Apart from this classical model, the MPI community has over the years added other communication models; however, their suitability for handling the irregular traffic workloads typical of graph operations remain comparatively less explored. Our aim in this paper is to study these relatively underutilized communication models of MPI for graph applications. More specifically, we evaluate MPI’s one-sided programming, or Remote Memory Access (RMA), and nearest neighborhood collectives using a process graph topology. There are features in these newer models that are intended to better map to irregular communication patterns, as exemplified in graph algorithms. As a concrete application for our case study, we use distributed memory implementations of a weighted graph matching algorithm to investigate performances of MPI-3 RMA and neighborhood collective operations compared to nonblocking Send- Recv. A matching in a graph is a subset of edges such that no two matched edges are incident on the same vertex. A maximum weight matching is a matching of maximum weight computed as the sum of the weights of matched edges. Algorithms for optimal matching have quadratic complexity and are difficult to parallelize. More importantly, execution of graph matching is dominated by high volume of irregular memory accesses, making it an ideal candidate for studying the effects of various MPI communication models on graph applications at scale. Our neighborhood collectives and RMA implementations yields up to 6× speedup over traditional optimized nonblocking Send-Recv implementations, on thousands of cores of the NERSC Cori supercomputer. We believe the lessons learned from this case study can be adopted to benefit a wider range of graph application codes.

Ghosh, Sayan↗

TBAA20: Task-Based Algorithms and Applications

The new challenges posed by Exascale system architectures have resulted in difficulty achieving a desired scalability using traditional distributed ­memory runtimes. Task­-based programming models show promise in addressing these challenges, providing application developers with a productive and performant approach to programming on next generation systems. Empirical studies show that task-based models can overcome load ­balancing issues that are inherent to traditional distributed ­memory runtimes, and that task-­based runtimes perform comparably to those systems when balanced. This panel is designed to explore the advantages of task-­based programming models on modern and future HPC systems from an industry, university, and national lab perspective. It aims at gathering application experts and proponents of these models to present concrete and practical examples of using task­-based runtimes to overcome the challenges posed by Exascale system architectures. This report describes the objectives, activities, and outcomes of the panel TBAA: Task­-Based Algorithms and Applications which was held at the International Conference for High Performance Computing, Networking, Storage, and Analysis (SC 20) on November 18, 2020.

97 MATHEMATICS AND COMPUTING↗

LDRD final report : massive multithreading applied to national infrastructure and informatics.

Large relational datasets such as national-scale social networks and power grids present different computational challenges than do physical simulations. Sandia's distributed-memory supercomputers are well suited for solving problems concerning the latter, but not the former. The reason is that problems such as pattern recognition and knowledge discovery on large networks are dominated by memory latency and not by computation. Furthermore, most memory requests in these applications are very small, and when the datasets are large, most requests miss the cache. The result is extremely low utilization. We are unlikely to be able to grow out of this problem with conventional architectures. As the power density of microprocessors has approached that of a nuclear reactor in the past two years, we have seen a leveling of Moores Law. Building larger and larger microprocessor-based supercomputers is not a solution for informatics and network infrastructure problems since the additional processors are utilized to only a tiny fraction of their capacity. An alternative solution is to use the paradigm of massive multithreading with a large shared memory. There is only one instance of this paradigm today: the Cray MTA-2. The proposal team has unique experience with and access to this machine. The XMT, which is now being delivered, is a Red Storm machine with up to 8192 multithreaded 'Threadstorm' processors and 128 TB of shared memory. For many years, the XMT will be the only way to address very large graph problems efficiently, and future generations of supercomputers will include multithreaded processors. Roughly 10 MTA processor can process a simple short paths problem in the time taken by the Gordon Bell Prize-nominated distributed memory code on 32,000 processors of Blue Gene/Light. We have developed algorithms and open-source software for the XMT, and have modified that software to run some of these algorithms on other multithreaded platforms such as the Sun Niagara and Opteron multi-core chips.

97 MATHEMATICS AND COMPUTING↗

pGraph: Efficient Parallel Construction of Large-Scale Protein Sequence Homology Graphs

Detecting sequence homology between protein sequences is a fundamental problem in computational molecular biology, with a pervasive application in nearly all analyses that aim to structurally and functionally characterize protein molecules. While detecting the homology between two protein sequences is relatively inexpensive, detecting pairwise homology for a large number of protein sequences can become computationally prohibitive for modern inputs, often requiring millions of CPU hours. Yet, there is currently no robust support to parallelize this kernel. In this paper, we identify the key characteristics that make this problemparticularly hard to parallelize, and then propose a new parallel algorithm that is suited for detecting homology on large data sets using distributed memory parallel computers. Our method, called pGraph, is a novel hybrid between the hierarchical multiple-master/worker model and producer-consumer model, and is designed to break the irregularities imposed by alignment computation and work generation. Experimental results show that pGraph achieves linear scaling on a 2,048 processor distributed memory cluster for a wide range of inputs ranging from as small as 20,000 sequences to 2,560,000 sequences. In addition to demonstrating strong scaling, we present an extensive report on the performance of the various system components and related parametric studies.

Environmental Molecular Sciences Laboratory↗

MDLoader: A Hybrid Model-Driven Data Loader for Distributed Graph Neural Network Training

Scalable data management is essential for processing large scientific dataset on HPC platforms for distributed deep learning. In-memory distributed storage is preferred for its speed, enabling rapid, random, and frequent data access required by stochastic optimizers. Processes use one-sided or collective communication to fetch remote data, with optimal performance depending on (i) dataset characteristics, (ii) training scale, and (iii) interconnection network. Empirical analysis shows collective communication excels with larger mini-batch sizes and/or fewer processes, whereas one-sided communication outperforms at larger scales. We propose MDLoader, a hybrid in-memory data loader for distributed graph neural network training. MDLoader features a model-driven performance estimator that dynamically selects between one-sided and collective communication at the beginning of training using Tree of Parzen Estimators (TPE). Evaluations on NERSC Perlmutter and OLCF Summit show MDLoader outperforms single-backend loaders by up to 2.83 × and predicts the suitable communication method with 96.3% (Perlmutter) and 94.3% (Summit) success rate.

Bae, Jonghyun↗

Design and performance of a scalable, parallel statistics toolkit.

Most statistical software packages implement a broad range of techniques but do so in an ad hoc fashion, leaving users who do not have a broad knowledge of statistics at a disadvantage since they may not understand all the implications of a given analysis or how to test the validity of results. These packages are also largely serial in nature, or target multicore architectures instead of distributed-memory systems, or provide only a small number of statistics in parallel. This paper surveys a collection of parallel implementations of statistics algorithm developed as part of a common framework over the last 3 years. The framework strategically groups modeling techniques with associated verification and validation techniques to make the underlying assumptions of the statistics more clear. Furthermore it employs a design pattern specifically targeted for distributed-memory parallelism, where architectural advances in large-scale high-performance computing have been focused. Moment-based statistics (which include descriptive, correlative, and multicorrelative statistics, principal component analysis (PCA), and k-means statistics) scale nearly linearly with the data set size and number of processes. Entropy-based statistics (which include order and contingency statistics) do not scale well when the data in question is continuous or quasi-diffuse but do scale well when the data is discrete and compact. We confirm and extend our earlier results by now establishing near-optimal scalability with up to 10,000 processes.

99 GENERAL AND MISCELLANEOUS↗

DeStager: feature guided in-situ data management in distributed deep memory hierarchies

In-situ analytics have been increasingly adopted by leadership scientific applications to gain fast insights into massive output data of simulations. With the current practice, systems buffer the output data in DRAM for analytics processing, constraining it to DRAM capacity un-used by the simulation. The rapid growth of data size requires alternative approaches to accommodating data-rich analytics, such as using solid-state disks to increase effective memory capacity. For this purpose, this paper explores software solutions for exploring the deep memory hierarchies expected on future high-end machines. Leveraging the fact that many analytics are sensitive to data features (regions-of-interest) hidden in the data being processed, the approach incorporates the knowledge of the data features into in-situ data management. It uses adaptive index creation/refinement to reduce the overhead of index management. In addition, it uses data features to predict data skew and improve load balance through controlling data distribution and placement on distributed staging servers. The experimental results show that such feature-guided optimizations achieve substantial improvements over state-of-the-art approaches for managing output data in-situ.

Computer Science↗

Tuple checkout with notify in coordination namespace system

A system and method for notifying a process about a creation or removal event of a named data element (NDE) in a coordination namespace distributed memory system. A controller runs methods to: generate a tuple corresponding to data generated by a requesting process, the tuple having a tuple name and data value; and generate a notification indicator in a pending notification list to indicate to one or more processes a notification of the creation or removal event associated with the corresponding tuple. Upon detecting the event performed on the tuple by a second process, the method further searches for NDEs in the distributed memory system having the same tuple name, and in response to determining an existence of an associated pending notification record in a pending notification list of records, notify each corresponding process of the one or more processes indicated in the list of the creation or removal event.

Jacob, Philip↗

Using Coarrays to Parallelize Legacy Fortran Applications: Strategy and Case Study

This paper summarizes a strategy for parallelizing a legacy Fortran 77 program using the object-oriented (OO) and coarray features that entered Fortran in the 2003 and 2008 standards, respectively. OO programming (OOP) facilitates the construction of an extensible suite of model-verification and performance tests that drive the development. Coarray parallel programming facilitates a rapid evolution from a serial application to a parallel application capable of running on multicore processors and many-core accelerators in shared and distributed memory. We delineate 17 code modernization steps used to refactor and parallelize the program and study the resulting performance. Our initial studies were done using the Intel Fortran compiler on a 32-core shared memory server. Scaling behavior was very poor, and profile analysis using TAU showed that the bottleneck in the performance was due to our implementation of a collective, sequential summation procedure. We were able to improve the scalability and achieve nearly linear speedup by replacing the sequential summation with a parallel, binary tree algorithm. We also tested the Cray compiler, which provides its own collective summation procedure. Intel provides no collective reductions. With Cray, the program shows linear speedup even in distributed-memory execution. We anticipate similar results with other compilers once they support the new collective procedures proposed for Fortran 2015.

Radhakrishnan, Hari↗

Compiler Optimization for Extreme-Scale Scripting

The data-driven task parallelism execution model can support parallel programming models that are well suited for large-scale distributed-memory parallel computing, for example, simulations and analysis pipelines running on clusters and clouds. We describe a novel compiler intermediate representation and optimizations for this execution model, including adaptions of standard techniques alongside novel techniques. These techniques are applied to Swift/T, a high-level scripting language for flexible data flow composition of functions, which may be serial or use lower-level parallel programming models such as MPI and OpenMP. This paper presents preliminary results, indicating that our compiler optimizations reduce communication overhead by 70% to 93% on distributed-memory systems.

Armstrong, Timothy G.↗

miniVite: A Graph Analytics Benchmarking Tool for Massively Parallel Systems

Benchmarking of high performance computing sys-tems can help provide critical insights for ef?cient design of computing systems and software applications. Although a large number of tools for benchmarking exist, there is a lack of rep-resentative benchmarks for the class of irregular computations as exempli?ed by graph analytics. In this paper, we propose miniVite as a representative graph analytics benchmark tool to test a variety of distributed-memory systems. Graph clustering, popularly known as community detection, is a prototypical graph operation used in numerous scienti?c computing and analytics applications. The goal of clustering is to partition a graph into clusters (or communities) such that each cluster consists of vertices that are densely connected within the cluster and sparsely connected to the rest of the graph. Modu-larity optimization is a popular technique for identifying clusters in a graph. Ef?cient parallelization of modularity optimization-based algorithms is challenging. One successful approach was conceived in Vite, a distributed-memory implementation of the Louvain algorithm that incorporates several heuristics. We introduce miniVite as a representative but simpli?ed variant of Vite, to serve as a prototypical graph analytics benchmarking tool. Unlike other graph-based methods such as breadth-?rst search and betweenness centrality, miniVite represents highly complex computational patterns stressing a variety of system features, which can provide crucial insight for co-design of future computing systems.

community detection, parallel, MPI↗

Avoiding Communication in Primal and Dual Block Coordinate Descent Methods

Primal and dual block coordinate descent methods are iterative methods for solving regularized and unregularized optimization problems. Distributed-memory parallel implementations of these methods have become popular in analyzing large machine learning datasets. However, existing implementations communicate at every iteration, which, on modern data center and supercomputing architectures, often dominates the cost of floating-point computation. Recent results on communication-avoiding Krylov subspace methods suggest that large speedups are possible by re-organizing iterative algorithms to avoid communication. We show how applying similar algorithmic transformations can lead to primal and dual block coordinate descent methods that only communicate every $s$ iterations---where $s$ is a tuning parameter---instead of every iteration for the regularized least-squares problem. We show that the communication-avoiding variants reduce the number of synchronizations by a factor of $s$ on distributed-memory parallel machines without altering the convergence rate and attain strong scaling speedups of up to $6.1\times$ over the “standard algorithm" on a Cray XC30 supercomputer.

97 MATHEMATICS AND COMPUTING↗

Distributed Macroscopic Traffic Simulation with Open Traffic Models

This paper presents OTM-MPI, an extension of the Open Traffic Models platform (OTM) for running macroscopic traffic simulations in high-performance computing environments. OTM-MPI represents the first open-source, distributed-memory, macroscopic simulation model developed for modern high performance parallel machines and large networks. Macroscopic simulations are appropriate for studying regional traffic scenarios when aggregate trends are of interest, rather than individual vehicle traces. They are also appropriate for studying the routing behavior of classes of vehicles, such as app-informed vehicles. The network partitioning was performed with METIS. Inter-process communication was done with MPI (message-passing interface). Results are provided for two networks: one realistic network which was obtained from Open Street Maps for Chattanooga, TN, and another larger synthetic grid network. The software recorded a speedups of 198x using 256 cores for Chattanooga, and 475x with 1,024 cores for the synthetic network.

macro-scopic traffic simulation↗

An Uncertainty Quantification Framework for Studying the Effect of Spatial Heterogeneity in Reservoir Permeability on CO2 Sequestration

In this paper, we introduce an uncertainty quantification (UQ) software framework for carbon sequestration, focused on the effect of spatial heterogeneity of reservoir properties on CO2 migration. We use a sequential Gaussian method (SGSIM) to generate realizations of permeability fields with various spatial statistical attributes. To deal with the computational difficulties, we integrate the following ideas/approaches. First, we use three different sampling approaches (probabilistic collocation, quasi-Monte Carlo, and adaptive sampling) to reduce the number of forward calculations while trying to explore the parameter space and quantify the input uncertainty. Second, we use eSTOMP as the forward modeling simulator. eSTOMP is implemented with the Global Arrays toolkit that is based on one-sided inter-processor communication and supports a shared memory programming style on distributed memory platforms, providing a highly-scalable performance. Third, we built an adaptive system infrastructure to select the best possible data transfer mechanisms, to optimally allocate system resources to improve performance and to integrate software packages and data for composing carbon sequestration simulation, computation, analysis, estimation and visualization. We demonstrate the framework with a given CO2 injection scenario in heterogeneous sandstone reservoirs.

Carbon sequestration, uncertainty quantification, ↗

Simulation and big data challenges in tuning building energy models

EnergyPlus is the flagship building energy simulation software used to model whole building energy consumption for residential and commercial establishments. A typical input to the program often has hundreds, sometimes thousands of parameters which are typically tweaked by a buildings expert to “get it right”. This process can sometimes take months. “Autotune” is an ongoing research effort employing machine learning techniques to automate the tuning of the input parameters for an EnergyPlus input description of a building. Even with automation, the computational challenge faced to run the tuning simulation ensemble is daunting and requires the use of supercomputers to make it tractable in time. In this paper, we describe the scope of the problem, particularly the technical challenges faced and overcome, and the software infrastructure developed/in development when taking the EnergyPlus engine, which was primarily designed to run on desktops, and scaling it to run on shared memory supercomputers (Nautilus) and distributed memory supercomputers (Frost and Titan). The parametric simulations produce data in the order of tens to a couple of hundred terabytes. We describe the approaches employed to streamline and reduce bottlenecks in the workflow for this data, which is subsequently being made available for the tuning effort as well as made available publicly for open-science.

Sanyal, Jibonananda↗

Simulation and Big Data Challenges in Tuning Building Energy Models

EnergyPlus is the flagship building energy simulation software used to model whole building energy consumption for residential and commercial establishments. A typical input to the program often has hundreds, sometimes thousands of parameters which are typically tweaked by a buildings expert to get it right . This process can sometimes take months. Autotune is an ongoing research effort employing machine learning techniques to automate the tuning of the input parameters for an EnergyPlus input description of a building. Even with automation, the computational challenge faced to run the tuning simulation ensemble is daunting and requires the use of supercomputers to make it tractable in time. In this proposal, we describe the scope of the problem, the technical challenges faced and overcome, the machine learning techniques developed and employed, and the software infrastructure developed/in development when taking the EnergyPlus engine, which was primarily designed to run on desktops, and scaling it to run on shared memory supercomputers (Nautilus) and distributed memory supercomputers (Frost and Titan). The parametric simulations produce data in the order of tens to a couple of hundred terabytes.We describe the approaches employed to streamline and reduce bottlenecks in the workflow for this data, which is subsequently being made available for the tuning effort as well as made available publicly for open-science.

Sanyal, Jibonananda↗

PLANC: Parallel Low-rank Approximation with Nonnegativity Constraints

In this work, we consider the problem of low-rank approximation of massive dense nonnegative tensor data, for example, to discover latent patterns in video and imaging applications. As the size of data sets grows, single workstations are hitting bottlenecks in both computation time and available memory. We propose a distributed-memory parallel computing solution to handle massive data sets, loading the input data across the memories of multiple nodes, and performing efficient and scalable parallel algorithms to compute the low-rank approximation. We present a software package called Parallel Low-rank Approximation with Nonnegativity Constraints, which implements our solution and allows for extension in terms of data (dense or sparse, matrices or tensors of any order), algorithm (e.g., from multiplicative updating techniques to alternating direction method of multipliers), and architecture (we exploit GPUs to accelerate the computation in this work). We describe our parallel distributions and algorithms, which are careful to avoid unnecessary communication and computation, show how to extend the software to include new algorithms and/or constraints, and report efficiency and scalability results for both synthetic and real-world data sets.

97 MATHEMATICS AND COMPUTING↗