Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel computer architecture”

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 559 records · Page 31

CDFG Extraction Tool for LLVM

With the dawn of the exascale era, computer scientists and engineers are faced with tremendous challenges across all facets of the HPC system - scalability, performance, reliability, and power consumption. In particular, the power-performance benefit from one processor generation to the next is seeing ever-diminishing returns and will require fundamental changes in the way we approach computation. In fact, it is likely that different applications will require different types of accelerators in order to meet power, performance, and reliability requirements at scale. One potential type of accelerator, a dataflow architecture, diverges from the traditional sequentially executed instruction model into one that reflects the inherent instruction-level parallelism in a program. This work presents the initial steps toward a tool that can extract the control-dataflow graph from an application.

97 MATHEMATICS AND COMPUTING↗

SimNet: Accurate and High-Performance Computer Architecture Simulation using Deep Learning

While cycle-accurate simulators are essential tools for architecture research, design, and development, their practicality is limited by an extremely long time-to-solution for realistic applications under investigation. This work describes a concerted effort, where machine learning (ML) is used to accelerate microarchitecture simulation. First, an ML-based instruction latency prediction framework that accounts for both static instruction properties and dynamic processor states is constructed. Then, a GPU-accelerated parallel simulator is implemented based on the proposed instruction latency predictor, and its simulation accuracy and throughput are validated and evaluated against a state-of-the-art simulator. Leveraging modern GPUs, the ML-based simulator outperforms traditional CPU-based simulators significantly.

97 MATHEMATICS AND COMPUTING↗

Snowmass Computational Frontier: Topical Group Report on Experimental Algorithm Parallelization

The substantial increase in data volume and complexity expected from future experiments will require significant investment to prepare experimental algorithms. These algorithms include physics object reconstruction, calibrations, and processing of observational data. In addition, the changing computing architecture landscape, which will be primarily composed of heterogeneous resources, will continue to pose major challenges with regard to algorithmic migration. Portable tools need to be developed that can be shared among the frontiers (e.g., for code execution on different platforms) and opportunities, such as forums or cross-experimental working groups, need to be provided where experiences and lessons learned can be shared between experiments and frontiers. At the same time, individual experiments also need to invest considerable resources to develop algorithms unique to their needs (e.g., for facilities dedicated to the experiment), and ensure that their specific algorithms will be able to efficiently exploit external heterogeneous computing facilities. Common software tools represent a cost-effective solution, providing ready-to-use software solutions as well as a platform for R&D work. These are particularly important for small experiments which typically do not have dedicated resources needed to face the challenges imposed by the evolving computing technologies. Workforce development is a key concern across frontiers and experiments, and additional support is needed to provide career opportunities for researchers working in the field of experimental algorithm development. Finally, cross-discipline collaborations going beyond high-energy physics are a key ingredient to address the challenges ahead and more support for such collaborations needs to be created. This report targets future experiments, observations and experimental algorithm development for the next 10-15 years.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Predicting the Electronic Properties of 3D, Million-atom Semiconductor nanostructure Architectures

This final report describes the work done by Jack Dongarra (University Distinguished Professor) and Stanimire Tomov (Research Scientist) related to the DOE project entitled Predicting the Electronic Properties of 3D, Million-Atom Semiconductor Nanostructure Architectures. In this project we addressed the mathematical methodology required to calculate the electronic and transport properties of large nanostructures with comparable accuracy and reliability to that of current ab initio methods. This capability is critical for further developing the field, yet it is missing in all the existing computational methods. Additionally, quantitative comparisons with experiments are often needed for a qualitative understanding of the physics, and for guiding the design of new nanostructures. We focused on the mathematical challenges of the project, in particular on solvers and preconditioners for large scale eigenvalue problems that occur in the computation of electronic states of large nanosystems. Usually, the states of interest lie in the interior of the spectrum and their computation poses great difficulties for existing algorithms. The electronic properties of a semiconductor nanostructure architecture can be predicted/determined by computing its band structure. Of particular importance are the 'band edge states' (electronic states near the energy gap) which can be computed from a properly defined interior eigenvalue problem. Our primary mathematics and computational challenge here has been to develop an efficient solution methodology for finding these interior states for very large systems. Our work has produced excellent results in terms of developing both new and extending current state-of-the-art techniques.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Enabling Co-Design of Multi-Layer Exascale Storage Architectures

Growing demands for computing power in applications such as energy production, climate analysis, computational chemistry, and bioinformatics have propelled computing systems toward the exascale: systems with 10 18 floating-point operations per second. These systems, to be designed and constructed over the next decade, will create unprecedented challenges in component counts, power consumption, resource limitations, and system complexity. Data storage and access are an increasingly important and complex component in extreme-scale computing systems, and significant design work is needed to develop successful storage hardware and software architectures at exascale. Co-design of these systems will be necessary to find the best possible design points for exascale systems. The goal of this work has been to enable the exploration and co-design of exascale storage systems by providing a detailed, accurate, and highly parallel simulation of exascale storage and the surrounding environment. Specifically, this simulation has (1) portrayed realistic application checkpointing and analysis workloads, (2) captured the complexity, scale, and multilayer nature of exascale storage hardware and software, and (3) executed in a timeframe that enables “what if'” exploration of design concepts. We developed models of the major hardware and software components in an exascale storage system, as well as the application I/O workloads that drive them. We used our simulation system to investigate critical questions in reliability and concurrency at exascale, helping guide the design of future exascale hardware and software architectures. Additionally, we provided this system to interested vendors and researchers so that others can explore the design space. We validated the capabilities of our simulation environment by configuring the simulation to represent the Argonne Leadership Computing Facility Blue Gene/Q system and comparing simulation results for application I/O patterns to the results of executions of these I/O kernels on the actual system.

97 MATHEMATICS AND COMPUTING↗

Petascale Computations for Large-Scale Atomic and Molecular Collisions. In: Sustained Simulations Performance.

Petaflop architectures are currently being utilized efficiently to perform large scale computations in Atomic, Molecular and Optical Collisions. We solve the Schrödinger or Dirac equation for the appropriate collision problem using the R-matrix or R-matrix with pseudo-states approach. We briefly outline the parallel methodology used and implemented for the current suite of Breit-Pauli and DARC codes. Various examples are shown of our theoretical results compared with those obtained from Synchrotron Radiation facilities and from Satellite observations. We also indicate future directions and implementation of the R-matrix codes on emerging GPU architectures.

Accurate Cross Section↗

Global Arrays Parallel Programming Toolkit

The two predominant classes of programming models for parallel computing are distributed memory and shared memory. Both shared memory and distributed memory models have advantages and shortcomings. Shared memory model is much easier to use but it ignores data locality/placement. Given the hierarchical nature of the memory subsystems in modern computers this characteristic can have a negative impact on performance and scalability. Careful code restructuring to increase data reuse and replacing fine grain load/stores with block access to shared data can address the problem and yield performance for shared memory that is competitive with message-passing. However, this performance comes at the cost of compromising the ease of use that the shared memory model advertises. Distributed memory models, such as message-passing or one-sided communication, offer performance and scalability but they are difficult to program. The Global Arrays toolkit attempts to offer the best features of both models. It implements a shared-memory programming model in which data locality is managed by the programmer. This management is achieved by calls to functions that transfer data between a global address space (a distributed array) and local storage. In this respect, the GA model has similarities to the distributed shared-memory models that provide an explicit acquire/release protocol. However, the GA model acknowledges that remote data is slower to access than local data and allows data locality to be specified by the programmer and hence managed. GA is related to the global address space languages such as UPC, Titanium, and, to a lesser extent, Co-Array Fortran. In addition, by providing a set of data-parallel operations, GA is also related to data-parallel languages such as HPF, ZPL, and Data Parallel C. However, the Global Array programming model is implemented as a library that works with most languages used for technical computing and does not rely on compiler technology for achieving parallel efficiency. It also supports a combination of task- and data-parallelism and is available as an extension of the message passing (MPI) model. The GA model exposes to the programmer the hierarchical memory of modern high-performance computer systems, and by recognizing the communication overhead for remote data transfer, it promotes data reuse and locality of reference. Virtually all the scalable architectures possess non-uniform memory access characteristics that reflect their multi-level memory hierarchies. These hierarchies typically comprise processor registers, multiple levels of cache, local memory, and remote memory. Over time, both the number of levels and the cost (in processor cycles) of accessing deeper levels has been increasing. It is important for any scalable programming model to address memory hierarchy since it is critical to the efficient execution of scalable applications.

ARMCI↗

Fast and Scalable Implementations of Influence Maximization Algorithms

The Influence Maximization problem has been extensively studied in the past decade because of its practical applications in finding the key influencers in social networks. Due to the hardness of the underlying problem, existing algorithms have tried to trade off practical efficiency with approximation guarantees. However, even such approximate solutions take several hours of compute time on modest sized real world inputs. On the other hand, there is a lack of effective parallel and distributed algorithms to solve this problem. In this paper, we present efficient parallel algorithms for multithreaded and distributed systems to solve the influence maximization with approximation guarantee. Our algorithms extend state-of-the-art sequential approach based on computing reverse reachability sets. We present a detailed experimental evaluation, and analyze their performance and their sensitivity to input parameters, using real world inputs. Our experimental results demonstrate significant speedup on parallel architectures. We further show a speedup of up to 586X relative to the state-of-the-art sequential baseline using 1024 nodes of a supercomputer at far greater accuracy and twice the seed set size. To the best of our knowledge, this is the first effort in parallelizing the influence maximization operation at scale.

Minutoli, Marco↗

Accelerating Binarized Neural Networks via Bit-Tensor-Cores in Turing GPUs

Despite foreseeing tremendous speedups over conventional deep neural networks, the performance advantage of binarized neural networks (BNNs) has merely been showcased on general-purpose processors such as CPUs and GPUs. In fact, due to being unable to leverage bit-level-parallelism with a word-based architecture, GPUs have been criticized for extremely low utilization (1%) when executing BNNs. Consequently, the latest tensorcores in NVIDIA Turing GPUs start to experimentally support bit computation. In this work, we look into this brand new bit computation capability and characterize its unique features. We show that the stride of memory access can significantly affect performance delivery and a data-format co-design is highly desired to support the tensorcores for achieving superior performance than existing software solutions without tensorcores. We realize the tensorcore-accelerated BNN design, particularly the major functions for fully-connect and convolution layers — bit matrix multiplication and bit convolution. Evaluations on two NVIDIA Turing GPUs show that, with ResNet-18, our BTC-BNN design can process ImageNet at a rate of 5.6K images per second, 77% faster than state-of-the-art. Our BNN approach is released on https://github.com/pnnl/TCBNN.

Li, Ang↗

From Physics Model to Results: An Optimizing Framework for Cross-Architecture Code Generation

Starting from a high-level problem description in terms of partial differential equations using abstract tensor notation, the Chemora framework discretizes, optimizes, and generates complete high performance codes for a wide range of compute architectures. Chemora extends the capabilities of Cactus, facilitating the usage of large-scale CPU/GPU systems in an efficient manner for complex applications, without low-level code tuning. Chemora achieves parallelism through MPI and multi-threading, combining OpenMP and CUDA. Optimizations include high-level code transformations, efficient loop traversal strategies, dynamically selected data and instruction cache usage strategies, and JIT compilation of GPU code tailored to the problem characteristics. The discretization is based on higher-order finite differences on multi-block domains. Chemora's capabilities are demonstrated by simulations of black hole collisions. This problem provides an acid test of the framework, as the Einstein equations contain hundreds of variables and thousands of terms.

Blazewicz, Marek↗

Towards Exascale Supernova Simulations with GenASiS

Core-collapse supernovae are worth simulating on leadership computing resources because they are the origin of many elements, are interesting targets for expensive observational programs, and provide valuable computational experience. The need to perform implicit neutrino transport addressing up to six dimensions of phase space makes simulations aimed at the explosion mechanism an exascale problem. The multiscale nature of the problem in space and time can be addressed with adaptive mesh refinement and implicit evolution. Several types of solvers - explicit and implicit, hyperbolic and possibly elliptic - must be deployed with operator splitting to address the multiphysics nature of the problem. While we have some ideas about how parallelism can be exploited, we look to applied mathematicians and computer scientists to work with us and provide libraries that deploy capabilities needed by our solvers on heterogeneous and communication-unfriendly computer architectures.

Cardall, Christian Y↗

The organic redox transistor for neuromorphic computing

Inspired by the in-memory computing architectures of biological systems, neuromorphic computing using crossbar arrays of artificial synapses based on non-volatile memory (NVM) devices with variable conductances has emerged as a new paradigm to enable massively parallel and ultra-low power computing hardware for data centric applications. Although inference has been demonstrated successfully using crossbars based on a variety of NMV technologies, efficient learning and scaling to large arrays (>10 6 elements) remains a challenge due to the synaptic elements' non-ideal electrical characteristics which degrades ANN accuracy. A further challenge is that in the conductive state memristors draw large currents >μA resulting in significant voltage drops in the interconnect wires and increased probability of failure in scaled arrays. We suggest the organic polymer redox transistor (RT) is an alternate approach that could solve many of these challenges, enabling both inference and parallel outer product updates, as recently demonstrated by Fuller et al. An RT consists of redox-active channel and gate electrodes in contact with a liquid or solid electrolyte. lon insertion through the electrolyte controls the channel electronic conductivity, while electron transfer through an external circuit maintains overall charge neutrality. Unlike a rechargeable battery, in the RT the voltage built-up across the electrolyte is kept to a minimum (typically <100 mV) by using the same material for the gate and channel. Elimination of the voltage offset simplifies integration of the RT into programmable arrays by enabling the use of various selectors. RTs based on inorganic and organic materials have been recently demonstrated with conductance tuning occurring at potentials of just a few mV and hundreds to thousands of linearly and symmetrically programmable conductance states, enabling near ideal accuracy in neural network simulations. Introduced in the 1980's, redox transistors with metallic gate electrodes and organic channel materials, also known as organic electrochemical transistors (OECTs), have been explored for a variety of applications such as chem- and bio-sensing, neural interfaces, and low cost printed circuits. A typical channel material for OECTs is the conducting polymer poly(3,4-ethylenedioxythiophene) doped with poly(styrene sulfonate) (PEDOT:PSS). PEDOT is a p-type semiconducting polymer with mobile positively charged polarons that hop chain-to-chain.

Talin, Albert Alec↗

Initial Characterization of Parallel NFS Implementations

Parallel NFS (pNFS) is touted as an emergent standard protocol for parallel I/O access in various storage environments. Several pNFS prototypes have been implemented for initial validation and protocol examination. Previous efforts have focused on realizing the pNFS protocol to expose the best bandwidth potential from underlying file and storage systems. In this presentation, we provide an initial characterization of two pNFS prototype implementations, lpNFS (a Lustre-based parallel NFS implementation) and spNFS (another reference implementation from Network Appliance, Inc.). We show that both lpNFS and spNFS can faithfully achieve the primary goal of pNFS, i.e., aggregating I/O bandwidth from many storage servers. However, they both face the challenge of scalable metadata management. Particularly, the throughput of sp-NFS metadata operations degrades significanlty with an increasing number of data servers. Even for the better-performing lpNFS, we discuss its architecture and propose a direct I/O request flow protocol to improve its performance.

97 MATHEMATICS AND COMPUTING↗

Fine-grained multithreading support for hybrid threaded MPI programming.

As high-end computing systems continue to grow in scale, recent advances in multi- and many-core architectures have pushed such growth toward more dense architectures, that is, more processing elements per physical node, rather than more physical nodes themselves. Although a large number of scientific applications have relied so far on an MPI-everywhere model for programming high-end parallel systems; this model may not be sufficient for future machines, given their physical constraints such as decreasing amounts of memory per processing element and shared caches. As a result, application and computer scientists are exploring alternative programming models that involve using MPI between address spaces and some other threaded model, such as OpenMP, Pthreads, or Intel TBB, within an address space. Such hybrid models require efficient support from an MPI implementation for MPI messages sent from multiple threads simultaneously. In this paper, we explore the issues involved in designing such an implementation. We present four approaches to building a fully thread-safe MPI implementation, with decreasing levels of critical-section granularity (from coarse-grain locks to fine-grain locks to lock-free operations) and correspondingly increasing levels of complexity. We present performance results that demonstrate the performance implications of the different approaches.

99 GENERAL AND MISCELLANEOUS↗

A parallel WLS state estimator on shared memory computers

This paper describes our experience with developing a parallel weighted- least-square (WLS) state estimation (SE) program for shared-memory parallel computers. Since the key computational kernel of the WLS algorithm based on the Newton-Raphson approach is a solver of sparse linear equations, a significant part of our effort was focused on selecting, implementing and evaluating this algorithm. An optimized shared memory version of the conjugate gradient (CG) algorithm was found to be competitive to state-of-the-art implementation of LU solvers for the SE problem on the SGI Altix, a shared memory architecture. We also ported the full SE algorithm including CG to the Cray MTA-2 shared memory multithreaded architecture and investigated its performance.

97 MATHEMATICS AND COMPUTING↗

Parallel computing for power system climate resiliency: Solving a large-scale stochastic capacity expansion problem with mpi-sppy

Here we propose a nodal stochastic generation and transmission expansion planning model that incorporates the output from high-resolution global climate models through load and generation availability scenarios. We implement our model in Pyomo and perform computational studies on a realistically-sized test case of the California electric grid in a high performance computing environment. We propose model reformulations and algorithm tuning to efficiently solve this large problem using a variant of the Progressive Hedging Algorithm. We utilize the parallelization capabilities and overall versatility of mpi-sppy, exploiting its hub-and-spoke architecture to concurrently obtain inner and outer bounds on an optimal expansion plan. Initial results show that instances with 360 representative days on a system with over 8,000 buses can be solved to within 5% of optimality in under 4 h of wall clock time, a first step towards solving a large-scale power system expansion planning problem across a wide range of climate-informed operational scenarios.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Unified Performance and Power Modeling of Scientific Workloads

It is expected that scientific applications executing on future large-scale HPC must be optimized not only in terms of performance, but also in terms of power consumption. As power and energy become increasingly constrained resources, researchers and developers must have access to tools that will allow for accurate prediction of both performance and power consumption. Reasoning about performance and power consumption in concert will be critical for achieving maximum utilization of limited resources on future HPC systems. To this end, we present a unified performance and power model for the Nek-Bone mini-application developed as part of the DOE's CESAR Exascale Co-Design Center. Our models consider the impact of computation, point-to-point communication, and collective communication

Modeling techniques↗

Maximal clique enumeration with data-parallel primitives

The enumeration of all maximal cliques in an undirected graph is a fundamental problem arising in several research areas. We consider maximal clique enumeration on shared-memory, multi-core architectures and introduce an approach consisting entirely of data-parallel operations, in an effort to achieve efficient and portable performance across different architectures. We study the performance of the algorithm via experiments varying over benchmark graphs and architectures. Overall, we observe that our algorithm achieves up to a 33-time speedup and 9-time speedup over state-of-the-art distributed and serial algorithms, respectively, for graphs with higher ratios of maximal cliques to total cliques. Further, we attain additional speedups on a GPU architecture, demonstrating the portable performance of our data-parallel design.

97 MATHEMATICS AND COMPUTING↗