Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Compilation”

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 1,747 records · Page 97

Utilizing many-core accelerators for halo and center finding within a cosmology simulation

Efficiently finding and computing statistics about “halos” (regions of high density) are essential analysis steps for N-body cosmology simulations. However, in state-of-the-art simulation codes, these analysis operators do not currently take advantage of the shared-memory data-parallelism available on multi-core and many-core architectures. The Hybrid / Hardware Accelerated Cosmology Code (HACC) is designed as an MPI+X code, but the analysis operators are parallelized only among MPI ranks, because of the difficulty in porting different X implementations (e.g., OpenMP, CUDA) across all architectures on which it is run. In this paper, we present portable data-parallel algorithms for several variations of halo finding and halo center finding algorithms. These are implemented with the PISTON component of the VTK-m framework, which uses Nvidia's Thrust library to construct data-parallel algorithms that allow a single implementation to be compiled to multiple backends to target a variety of multi-core and many-core architectures. Finally, we compare the performance of our halo and center finding algorithms against the original HACC implementations on the Moonlight, Stampede, and Titan supercomputers. The portability of Thrust allowed the same code to run efficiently on each of these architectures. On Titan, the performance improvements using our code have enabled halo analysis to be performed on a very large data set (81923 particles across 16,384 nodes of Titan) for which analysis using only the existing CPU algorithms was not feasible.

Sewell, Christopher↗

In-Memory Graph Databases for Web-Scale Data

RDF databases have emerged as one of the most relevant way for organizing, integrating, and managing expo- nentially growing, often heterogeneous, and not rigidly structured data for a variety of scientific and commercial fields. In this paper we discuss the solutions integrated in GEMS (Graph database Engine for Multithreaded Systems), a software framework for implementing RDF databases on commodity, distributed-memory high-performance clusters. Unlike the majority of current RDF databases, GEMS has been designed from the ground up to primarily employ graph-based methods. This is reflected in all the layers of its stack. The GEMS framework is composed of: a SPARQL-to-C++ compiler, a library of data structures and related methods to access and modify them, and a custom runtime providing lightweight software multithreading, network messages aggregation and a partitioned global address space. We provide an overview of the framework, detailing its component and how they have been closely designed and customized to address issues of graph methods applied to large-scale datasets on clusters. We discuss in details the principles that enable automatic translation of the queries (expressed in SPARQL, the query language of choice for RDF databases) to graph methods, and identify differences with respect to other RDF databases.

GEMS↗

PyOMP: Multithreaded Parallel Programming in Python

We know that Python is a widely used language in scientific computing. When the goal is high performance, however, Python lags far behind low-level languages such as C and Fortran. To support applications that stress performance, Python needs to access the full capabilities of modern CPUs. That means support for parallel multithreading. In this paper, we describe PyOMP, a system that enables OpenMP in Python. Programmers write code in Python with OpenMP, Numba generates code that compiles to LLVM, and the resulting programs run with performance that approaches that from code written with C and OpenMP. In this paper we provide an update on the PyOMP project and explain how to install it and use it to write parallel multithreaded code in Python.

97 MATHEMATICS AND COMPUTING↗

ICED: An Integrated CGRA Framework Enabling DFVS-Aware Acceleration

oarse-grained reconfigurable arrays (CGRAs) are a promising solution to enable energy-efficient acceleration of applications from different domains. By leveraging reconfiguration at the functional level, they can adapt to significantly different computational patterns. Existing CGRA mapping approaches extract instruction-level parallelism, exploit loop-pipelining opportunities, guarantee the data dependency, and target high throughput of a given loop. However, the recurrence data-dependency in the DFG and the mismatch between required and available computing/communication resources complicate the mapping, and might lead to significant unbalances in the utilization of the CGRA's tiles. This results in wasted power for tiles with low utilization. Applying dynamic voltage and frequency scaling (DVFS) can potentially solve this challenge and improve energy efficiency by adjusting voltage and frequency of different tiles independently. CGRAs have also been successful in accelerating data-dependent streaming applications. However, in these applications, the execution time of each kernel in the pipeline might dynamically vary depending on the characteristics of the input. This also leads to under-utilization of resources for the dynamically changing kernels that do not limit the application throughput. DVFS can also improve energy efficiency for these applications by dynamically changing the voltage and frequency levels of tiles that host non performance-constraining kernels. This paper proposes ICEDTEA -- an integrated DVFS-aware framework to map applications on CGRAs that support power islands. ICEDTEA proposes a CGRA architecture supporting DVFS islands at varying granularity (from a single tile to a group of tiles) and the related DVFS-aware compilation and mapping toolchain. ICEDTEA is the first work that introduces DVFS support for spatio-temporal CGRAs at power-island levels. The experimental evaluation shows that ICEDTEA improves average utilization by 2.3$\times$ and energy-efficiency by 1.32$\times$ over a conventional CGRA. With streaming applications, ICEDTEA improves energy efficiency by 1.12$\times$ over a state-of-the-art CGRA that introduces partial dynamic reconfiguration to adapt to variations in kernels' throughput.

Tan, Cheng↗

Intelligent Buildings of the Future: Cyberaware, Deep Learning Powered, and Human Interacting

This study discuss, intelligent buildings are quickly becoming cohesive and integral entities of cyber-physical ecosystems. Modern buildings adapt to internal and external elements and thrive on ever increasing data sources such as ubiquitous smart devices and sensors while mimicking various approaches previously known in software, hardware, and bio inspired systems. This article provides an overview of intelligent buildings of the future from a range of perspectives. First, a brief overview of prospects of U.S. and world energy consumption, concepts of smart buildings as integral entities of smart grids is presented. This is followed by a discussion of the role of connectivity and interoperability as well as current issues and solutions in the areas of security, resilience, and humans in realm of intelligent buildings. The article dives into detail about the intelligence aspect of intelligent buildings of the future. The learning and predictive Artificial Intelligence (AI) techniques behind many of the concepts of intelligent buildings are reviewed, including the latest techniques such as deep learning. As a concrete application of AI in the realm of intelligent buildings, prediction of energy consumption/load/demand, which is a much researched application of AI in the realm of intelligent buildings, is elaborated. The discussion is focused on most recent advancements of AI based demand prediction and the overview consists of existing research and a case study conducted by the authors for deep learning based demand prediction. Then, the article discusses the role of humans in intelligent buildings. Different types of human-building interactions are identified. Further, the article presents how each type of human-building interaction can benefit the overall control of intelligent buildings. The section is concluded by elaborating a case study conducted by the authors to achieve effective and efficient human-building interaction. The article is concluded with compilation of future insights based on latest technological advancements in U.S. industry and government.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Scaling Semantic Graph Databases in Size and Performance

In this paper we present SGEM, a full software system for accelerating large-scale semantic graph databases on commodity clusters. Unlike current approaches, SGEM addresses semantic graph databases by only employing graph methods at all the levels of the stack. On one hand, this allows exploiting the space efficiency of graph data structures and the inherent parallelism of graph algorithms. These features adapt well to the increasing system memory and core counts of modern commodity clusters. On the other hand, however, these systems are optimized for regular computation and batched data transfers, while graph methods usually are irregular and generate fine-grained data accesses with poor spatial and temporal locality. Our framework comprises a SPARQL to data parallel C compiler, a library of parallel graph methods and a custom, multithreaded runtime system. We introduce our stack, motivate its advantages with respect to other solutions and show how we solved the challenges posed by irregular behaviors. We present the result of our software stack on the Berlin SPARQL benchmarks with datasets up to 10 billion triples (a triple corresponds to a graph edge), demonstrating scaling in dataset size and in performance as more nodes are added to the cluster.

Aggregation↗

Data Layout Transformation for Enhancing Data Locality on NUCA Chip Multiprocessors

With increasing numbers of cores, future CMPs (Chip Multi-Processors) are likely to have a tiled architecture with a portion of shared L2 cache on each tile and a bankinterleaved distribution of the address space. Although such an organization is effective for avoiding access hot-spots, it can cause a significant number of non-local L2 accesses for many commonly occurring regular data access patterns. In this paper we develop a compile-time framework for data locality optimization via data layout transformation. Using a polyhedral model, the program’s localizability is determined by analysis of its index set and array reference functions, followed by non-canonical data layout transformation to reduce non-local accesses for localizable computations. Simulation-based results on a 16-core 2D tiled CMP demonstrate the effectiveness of the approach. The developed program transformation technique is also useful in several other data layout transformation contexts.

99 GENERAL AND MISCELLANEOUS↗

The Case for Custom TMY's: Examples Using the NSRDB

A typical meteorological year (TMY) data set essentially represents an hourly compilation of median months constructed using multi-year datasets. Although TMY data sets are generated from irradiance data in the horizontal plane, they are used in photovoltaic (PV) modeling for systems inclined to various angles. This paper demonstrates that a POA TMY (plane of array TMY) generated by selecting median months from a multi-year POA irradiance timeseries dataset produces significantly different results than a POA TMY generated by transposing a TMY dataset constructed from horizontal data. In some months the differences can be more than 3%. These results point to the need for generating TMY's using POA irradiance timeseries representing the orientation at which PV panels will be deployed.

14 SOLAR ENERGY↗

Statistical Performance Analysis on Approximately 320 Perovskite Single- and Two-Junction Solar Cells and Modules from >30 Global Sources

As perovskite photovoltaics (PV) advance from the laboratory to commercial prototypes, their accurate and reliable performance testing is becoming increasingly important. The well documented dynamic response of perovskite solar cells to an external applied voltage has led to the development of steady-state performance measurement methods; however, these methods have not been widely adopted by the perovskite PV community. A key reason for this is that steady-state measurement methods take tens of minutes to complete, as opposed to conventional "fast" current-voltage (I-V) measurements usually lasting a few seconds. Fast I-Vs arise from a snapshot, almost always not a steady-state condition of the device; however, given their widespread use, the question arises: how do performance parameters of perovskite PV compare when measured with fast I-V and with a steady-state method? We compile results from ca. 320 perovskite PV cells and modules, including single junction, and two-terminal perovskite-perovskite and perovskite-Si tandems, and show that fast I-Vs can provide a useful measure of the open-circuit voltage of the devices, while the short-circuit current and the overall efficiency can be widely misestimated. We discuss implications of these findings on performance testing protocols and propose possible options for fast and accurate testing of perovskite PV.

asymptotic↗

Optimization of Simultaneous Measurement for Variational Quantum Eigensolver Applications

Variational quantum eigensolver (VQE) is a promising algorithm suitable for near-term quantum computers. VQE aims to approximate solutions to exponentially-sized optimization problems by executing a polynomial number of quantum sub-problems. However, the number of subproblems scales as N-4 for typical problems of interest-a daunting growth rate that poses a serious limitation for emerging applications such as quantum computational chemistry. We mitigate this issue by exploiting the simultaneous measurability of subproblems corresponding to commuting terms. Our technique transpiles VQE instances into a format optimized for simultaneous measurement, ultimately yielding 8-30x lower cost. Our work also encompasses a synthesis tool for compiling simultaneous measurement circuits with minimal overhead. We demonstrate experimental validation of our techniques by estimating the ground state energy of deuteron with a quantum computer. We also investigate the underlying statistics of simultaneous measurement and devise an adaptive strategy for mitigating harmful covariance terms.

quantum computing↗

Numerical Simulations of Noisy Variational Quantum Eigensolver Ansatz Circuits

This is a case study of the variational quantum eigensolver (VQE) method using numerical simulations to test the influence of noise on the accuracy of the underlying circuit ansatz. We investigate a computational chemistry application of VQE to calculate the electronic ground state and its energy for Sodium Hydride (NaH), a prototypical two-electron problem. Using a one-parameter ansatz derived from unitary coupled cluster (UCC) theory, we simulate the effects of noise on the energy expectation value and variance with respect to the ansatz parameter. These numerical simulations provide insights into the accuracy of the prepared quantum state and the efficiency of the classical optimizer that iteratively refines the ansatz. We conduct a comparative study between analytical results derived for the UCC ansatz in the absence of noise and the noisy numerical simulation results obtained using an isotropic depolarizing noise model for each gate. We also compare the relative increase in noise on logically equivalent UCC ansatz circuits generated by randomized compiling. Notably, we observe that the intrinsic variance in the energy due to the simplicity of the ansatz itself compares with the noise induced by the bare circuit.

Gowrishankar, Meenambika↗

Adaptive job and resource management for the growing quantum cloud

As the popularity of quantum computing continues to grow, efficient quantum machine access over the cloud is critical to both academic and industry researchers across the globe. And as cloud quantum computing demands increase exponentially, the analysis of resource consumption and execution characteristics are key to efficient management of jobs and resources at both the vendor-end as well as the client-end. While the analysis and optimization of job / resource consumption and management are popular in the classical HPC domain, it is severely lacking for more nascent technology like quantum computing.This paper proposes optimized adaptive job scheduling to the quantum cloud taking note of primary characteristics such as queuing times and fidelity trends across machines, as well as other characteristics such as quality of service guarantees and machine calibration constraints. Key components of the proposal include a) a prediction model which predicts fidelity trends across machine based on compiled circuit features such as circuit depth and different forms of errors, as well as b) queuing time prediction for each machine based on execution time estimations. Altogether, this proposal is evaluated on simulated IBM machines across a diverse set of quantum applications and system loading scenarios, and is able to reduce wait times by over 3x and improve fidelity by over 40% on specific usecases, when compared to traditional job schedulers.

97 MATHEMATICS AND COMPUTING↗

Molecular Symmetry in VQE: A Dual Approach for Trapped-Ion Simulations of Benzene

Understanding complex chemical systems—such as biomolecules, catalysts, and novel materials—is a central goal of quantum simulations. Near-term strategies hinge on the use of variational quantum eigensolver (VQE) algorithms combined with a suitable ansatz. However, straightforward application of many chemically-inspired ansatze yields prohibitively deep circuits. In this work, we employ several circuit optimization methods tailored for trapped-ion quantum devices to enhance the feasibility of intricate chemical simulations. The techniques aim to lessen the depth of the unitary coupled cluster with singles and doubles (uCCSD) ansatz's circuit compilation, a considerable challenge on current noisy quantum devices. Furthermore, we use symmetry-inspired classical post-selection methods to further refine the outcomes and minimize errors in energy measurements, without adding quantum overhead. Our strategies encompass optimal mapping from orbital to qubit, term reordering to minimize entangling gates, and the exploitation of molecular spin and point group symmetry to eliminate redundant parameters. The inclusion of error mitigation via post-selection based on known molecular symmetries improves the results to near milli-Hartree accuracy. These methods, when applied to a benzene molecule simulation, enabled the construction of an 8-qubit circuit with 69 two-qubit entangling operations, pushing the limits for variational quantum eigensolver (VQE) circuits executed on quantum hardware to date. 1 1 This manuscript has been authored in part by UT-Battelle, LLC, under contract DE-AC05-000R22725 with the US Department of Energy (DOE). The publisher acknowledges the US government license to provide public access under the DOE Public Access Plan(https://energy.gov/doe-public-access-plan).

Goings, Joshua↗

Real-time performance of commercial Intel-based VME Controllers for the CODA data acquisition system

We have evaluated the performance of several Intel-based VME Controllers for the use in the data acquisition systems (DAQs) at Jefferson Lab. For the 12 GeV upgrade, PPC-based VME controllers running vxWorks will be replaced with those that are Intel-based running Linux. Their task will be facilitated by the use of FPGAs on the VME modules to perform trigger logic and communicating trigger information over serial and fiber connections through the DAQ. The need for a hard realtime operating system on the VME controller is removed from the equation as the readout of the digitized data from the VME modules (using VME-2eSST) is done in a threaded environment with multiple CPU cores while digitization takes place in the buffered, pipelined system. In this paper we brief discuss the 12 GeV Hall D DAQ and the requirements of the VME Controller. We present results from baseline testing of various models from different vendors using different Linux kernels, including results from a kernel compiled with the CONFIG_PREEMPT_RT patch.

Moffit, Bryan↗

Global Data Re-allocation via Communication Aggregation in Chapel

Chapel is a parallel programming language designed to improve the productivity and ease of use of conventional and parallel computers. This language currently delivers sub optimal performance when executing codes that perform global data re-allocation operations on distributed memory architectures. This is mainly due to data communication that is done without aggregation (one message for each remote array element). In this work, we analyze Chapel's standard Block and Cyclic distribution modules and optimize the communication routines for array assignments by performing aggregation. Thanks to the expressive power of Chapel, the compiler and runtime have enough information to do communication aggregation without user intervention. The runtime relies on the low-level GAS Net networking layer, whose versions of one-sided bulk put/get routines that support strides are particularly useful for us. Experimental results conducted on Hector (a Cray XE6) and Jaguar (Cray XK6)reveal that the implemented techniques can lead to significant reductions in communication time.

Sanz, Alberto↗

IRIS-GNN: Leveraging Graph Neural Networks for Scheduling on Truly Heterogeneous Runtime Systems

The diversity of accelerators in computer systems poses significant challenges for software developers, such as managing vendor-specific compiler toolchains, code fragmentation requiring different kernel implementations, and performance portability issues. To address these, the Intelligent Runtime System (IRIS) was developed. IRIS works across various systems, from smartphones to supercomputers, enabling automatic performance scaling based on available accelerators. It introduces abstract tasks for seamless execution transitions between accelerators while ensuring memory consistency and task dependencies. Although IRIS simplifies system details, optimal dynamic scheduling still requires user input to understand workload structures. To address this, we introduce a new scheduling policy for IRIS, termed IRIS-GNN, which is the first IRIS hybrid policy that operates in conjunction with the dynamic policies. This policy employs a Graph-Neural Network (GNN) to conduct Graph Classification of any task graphs submitted to IRIS. This GNN analyzes the structure and attributes of the task graph, categorizing it as either locality, concurrency, or mixed. This classification subsequently guides the selection of the dynamic policy used by IRIS. We provide a comparison of the performance of IRIS-GNN against the complete spectrum of IRIS’s dynamic policies, assess the overhead introduced by the GNN within this scheduling framework, and ultimately explore its practical application in real-world scenarios.

Johnston, Beau↗

A Dynamically Reconfigurable System for Closed-Loop Measurements of Network Traffic

Streaming network traffic measurement and analysis is critical for detecting and preventing any real-time anomalies in the network. The high speeds and complexity of today's networks, coupled with ever evolving threats, necessitate closing of the loop between measurements and their analysis in real time. The ensuing system demands high levels of programmability and processing where streaming measurements adapt to the changing network behavior in a goal-oriented manner. In this work, we exploit the features and requirements of the problem and develop an application-specific FPGA-based closed-loop measurement (CLM) system. We make novel use of fine-grained partial dynamic reconfiguration (PDR) as underlying reprogramming paradigm, performing low-latency just-in-time compiled logic changes in FPGA fabric corresponding to the dynamic measurement requirements. Our innovative dynamically reconfigurable socket offers 3× logic savings over conventional static solutions, while offering much reduced reconfiguration latencies over conventional PDR mechanisms. Here, we integrate multiple sockets in a highly parallel CLM framework and demonstrate its effectiveness in identifying heavy flows in streaming network traffic. The results using an FPGA prototype offer 100 percent detection accuracy while sustaining increasing link speeds.

97 MATHEMATICS AND COMPUTING↗

PANTHER: A Programmable Architecture for Neural Network Training Harnessing Energy-Efficient ReRAM

The wide adoption of deep neural networks has been accompanied by ever-increasing energy and performance demands due to the expensive nature of training them. Additionally, numerous special-purpose architectures have been proposed to accelerate training: both digital and hybrid digital-analog using resistive RAM (ReRAM) crossbars. ReRAM-based accelerators have demonstrated the effectiveness of ReRAM crossbars at performing matrix-vector multiplication operations that are prevalent in training. However, they still suffer from inefficiency due to the use of serial reads and writes for performing the weight gradient and update step. A few works have demonstrated the possibility of performing outer products in crossbars, which can be used to realize the weight gradient and update step without the use of serial reads and writes. However, these works have been limited to low precision operations which are not sufficient for typical training workloads. Moreover, they have been confined to a limited set of training algorithms for fully-connected layers only. To address these limitations, we propose a bit-slicing technique for enhancing the precision of ReRAM-based outer products, which is substantially different from bit-slicing for matrix-vector multiplication only. We incorporate this technique into a crossbar architecture with three variants catered to different training algorithms. To evaluate our design on different types of layers in neural networks (fully-connected, convolutional, etc.) and training algorithms, we develop PANTHER, an ISA-programmable training accelerator with compiler support. Our design can also be integrated into other accelerators in the literature to enhance their efficiency. Our evaluation shows that PANTHER achieves up to 8.02×, 54.21×, and 103× energy reductions as well as 7.16×, 4.02×, and 16× execution time reductions compared to digital accelerators, ReRAM-based accelerators, and GPUs, respectively.

42 ENGINEERING↗