Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “asynchronous algorithms”

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 145 records · Page 8

Decentralized Schemes with Overlap for Solving Graph-Structured Optimization Problems

We present a new algorithmic paradigm for the decentralized solution of graph-structured optimization problems that arise in the estimation and control of network systems. A key and novel design concept of the proposed approach is that it uses overlapping subdomains to promote and accelerate convergence. We show that the algorithm converges if the size of the overlap is sufficiently large and that the convergence rate improves exponentially with the size of the overlap. The proposed approach provides a bridge between fully decentralized and centralized architectures and is flexible in that it enables the implementation of asynchronous schemes, handling of constraints, and balancing of computing, communication, and data privacy needs. The proposed scheme is tested in an estimation problem for a 9241-node power network and we show that it outperforms the alternating direction method of multipliers.

asynchronous↗

Accelerating DCA++ (Dynamical Cluster Approximation) Scientific Application on the Summit supercomputer

Optimizing scientific applications on today’s accelerator-based high performance computing systems can be challenging, especially when multiple GPUs and CPUs with heterogeneous memories and persistent non-volatile memories are present. An example is Summit, an accelerator-based system at the Oak Ridge Leadership Computing Facility (OLCF) that is rated as the world’s fastest supercomputer to-date. New strategies are thus needed to expose the parallelism in legacy applications, while being amenable to efficient mapping to the underlying architecture.In this paper we discuss our experiences and strategies to port a scientific application, DCA++, to Summit. DCA++ is a highperformance research application that solves quantum manybody problems with a cutting edge quantum cluster algorithm, the dynamical cluster approximation.Our strategies aim to synergize the strengths of the different programming models in the code. These include: (a) streamlining the interactions between the CPU threads and the GPUs, (b) implementing computing kernels on the GPUs and decreasing CPU-GPU memory transfers, (c) allowing asynchronous GPU communications, and (d) increasing compute intensity by combining linear algebraic operations.Full-scale production runs using all 4600 Summit nodes attained a peak performance of 73.5 PFLOPS with a mixed precision implementation.We observed a perfect strong and weak scaling for the quantum Monte Carlo solver in DCA++, while encountering about 2× input/output (I/O) and MPI communication overhead on the time-to-solution for the full machine run. Our hardware agnostic optimizations are designed to alleviate the communication and I/O challenges observed, while improving the compute intensity and obtaining optimal performance on a complex, hybrid architecture like Summit.

Balduzzi, Giovanni↗

A Hartree-Fock Application Using UPC++ and the New DArray Library

The Hartree-Fock (HF) method is the fundamental first step for incorporating quantum mechanics into many-electron simulations of atoms and molecules, and it is an important component of computational chemistry toolkits like NWChem. The GTFock code is an HF implementation that, while it does not have all the features in NWChem, represents crucial algorithmic advances that reduce communication and improve load balance by doing an up-front static partitioning of tasks, followed by work stealing whenever necessary. To enable innovations in algorithms and exploit next generation exascale systems, it is crucial to support quantum chemistry codes using expressive and convenient programming models and runtime systems that are also efficient and scalable. Here, this paper presents an HF implementation similar to GTFock using UPC++, a partitioned global address space model that includes flexible communication, asynchronous remote computation, and a powerful multidimensional array library. UPC++ offers runtime features that are useful for HF such as active messages, a rich calculus for array operations, hardware-supported fetch-and-add, and functions for ensuring asynchronous runtime progress. We present a new distributed array abstraction, DArray, that is convenient for the kinds of random-access array updates and linear algebra operations on block-distributed arrays with irregular data ownership. Finally, we analyze the performance of atomic fetch-and-add operations (relevant for load balancing) and runtime attentiveness, then compare various techniques and optimizations for each. Our optimized implementation of HF using UPC++ and the DArrays library shows up to 20% improvement over GTFock with Global Arrays at scales up to 24,000 cores.

97 MATHEMATICS AND COMPUTING↗

Agent-Supervisor Coordination for Decentralized Event-Triggered Optimization

This letter proposes decentralized resource-aware coordination schemes for solving network optimization problems defined by objective functions that combine locally evaluable costs with network-wide coupling components. These methods are well suited for a group of supervised agents trying to solve an optimization problem under mild coordination requirements. Each agent has information on its local cost and coordinates with the network supervisor for information about the coupling term of the cost. The proposed approach is feedback-based and asynchronous by design, guarantees anytime feasibility, and ensures the asymptotic convergence of the network state to the desired optimizer. Numerical simulations on a power system example illustrate our results.

decentralized algorithms↗

Radiation-Induced Noise Resilience of Neuromorphic Architectures

Neuromorphic event-based networks use asynchronous time-dependent information to extract features from input data that can allow for edge-based distributed applications such as object recognition. The noise resilience properties of such networks, especially in the context of space applications, are yet to be explored. In this paper, we use the hierarchy of time surfaces (HOTS) algorithm, which is one of the neuromorphic algorithms, to understand the least and most resilient modules in a neuromorphic network. The HOTS algorithm relies on the computing of time surfaces that maps the temporal delays between neighboring pixels into normalized features that involve many computations that are also found in other neuromorphic networks such as exponential decays, distance computations, etcetera. We implemented HOTS on a Digilent PYNQ board with a Xilinx Zynq 7020 system on a chip, and we subjected the boards running the HOTS network inference to neutron radiation at the Los Alamos Neutron Science Center. Furthermore, we used simulation models from our previous similar experiments on the event-based sensor to create a neutron induced noise model to quantify the effect of this noise on the overall performance of the network. This experiment provides the preliminary measurements of the reliability of the HOTS algorithm and proposes methods to create a more reliable HOTS architecture in future spacecraft missions.

Engineering↗

Network Offoaded Hierarchical Collectives Using ConnectX-2's CORE-Direct Capabilities

As the scale of High Performance Computing (HPC) systems continues to increase, demanding that we extract even more parallelism from applications, the need to move communication management away from the Central Processing Unit (CPU) becomes even greater. Moving this management to the network, frees up CPU cycles for computation, making it possible to overlap computation and communication. In this paper we continue to investigate how to best use the new CORE-Direct support added in the ConnectX-2 Host Channel Adapter (HCA) for creating high performance, asynchronous collective operations that are managed by the HCA. Specifically we consider the network topology, creating a two-level communication hierarchy, reducing the MPI Barrier completion time by 45%, from 26.59 microseconds, when not considering network topology, to 14.72 microseconds, with the CPU based collective barrier operation completing in 19.04 microseconds. The nonblocking barrier algorithm has similar performance, with about 50% of that time available for computation.

99 GENERAL AND MISCELLANEOUS↗

Models and methods for assessing the value of HVDC and MVDC technologies in modern power grids

This report reflects the results of U.S. Department of Energy’s (DOE) Grid Modernization project 0074 “Models and methods for assessing the value of HVDC [high-voltage direct current] and MTDC [multi-terminal direct current] technologies in modern power grids.” The work was done by the Pacific Northwest National Laboratory (PNNL) and Oak Ridge National Laboratory (ORNL) in cooperation with Mid-Continent Independent System Operator (MISO) and Siemens. The main motivation of this study was to show the benefit of using direct current (DC) systems larger than those in existence today as they overlap with the alternating current (AC) systems. Proper use of their flexibility in terms of active/reactive power control and fast response can provide much-needed services to the grid at the same time as moving large blocks of energy to take advantage of cost diversity. Ultimately, the project’s success will enable decision-makers and investors to make well-informed decisions regarding this use of DC systems. This project showed the technical feasibility of HVDC macrogrid for frequency control and congestion relief in addition to bulk power transfers. Industry-established models for commonly used technologies were employed, along with high-fidelity models for recently developed HVDC converter technologies; like the modular multilevel converters (MMCs), a voltage source converters (VSC). Models for General Electric Positive Sequence Load Flow (GE PSLF) and Siemens Power System Simulator (PSS/E), widely used analysis programs, were for the first time adapted to include at the same time both Western Electricity Coordinating Council (WECC) and Eastern Interconnection (EI), the two largest North American interconnections. The high-fidelity models and their control were developed in detail for MMC system and extended to HVDC systems in point-to-point and in three-node multi-terminal configurations. Using a continental-level mixed AC-DC grid model, and using a HVDC macrogrid power flow and transient stability model, the results showed that the HVDC macrogrid relieved congestion and mitigated loop flows in AC networks, and provided up to 24% improvement in frequency responses. These are realistic studies, based on the 2025 heavy summer and EI multi-regional modeling working group (MMWG) 2026 summer peak cases. This work developed high-fidelity models and simulation algorithms to understand the dynamics of MMC. The developed models and simulation algorithms are up to 25 times faster than the existing algorithms. Models and control algorithms for high-fidelity models were designed and tested for point-to-point and multi-terminal configurations. The multi-terminal configuration was tested connecting simplified models of EI, WI, and Electric Reliability Council of Texas (ERCOT). The developed models showed up to 45% improvement in frequency response with the connection of all the three asynchronous interconnections in the United States using fast and advanced DC technologies like the multi-terminal MMC-DC system. Future work will look into developing high-fidelity models of other advanced DC technologies, combining high-fidelity models with the continental-level model, incorporating additional services. More scenarios involving large-scale HVDC and MTDC will be evaluated.

24 POWER TRANSMISSION AND DISTRIBUTION↗

GraphReduce: Processing Large-Scale Graphs on Accelerator-Based Systems

Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device’s internal memory capacity. GraphReduce adopts a combination of edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and device.

Architecture optimization, reuse, performance, ene↗

GraphReduce: Large-Scale Graph Analytics on Accelerator-Based HPC Systems

Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device’s internal memory capacity. GraphReduce adopts a combination of both edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and the device.

Architecture optimization, reuse, performance, ene↗

Algorithms for high-throughput disk-to-disk sorting, In: SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

In this paper, we present a new out-of-core sort algorithm, designed for problems that are too large to fit into the aggregate RAM available on modern supercomputers. We analyze the performance including the cost of IO and demonstrate the fastest (to the best of our knowledge) reported throughput using the canonical sortBenchmark on a general-purpose, production HPC resource running Lustre. By clever use of available storage and a formulation of asynchronous data transfer mechanisms, we are able to almost completely hide the computation (sorting) behind the IO latency. This latency hiding enables us to achieve comparable execution times, including the additional temporary IO required, between a large sort problem (5TB) run as a single, in-RAM sort and our out-of-core approach using 1/10th the amount of RAM. In our largest run, sorting 100TB of records using 1792 hosts, we achieved an end-to-end throughput of 1.24TB/min using our general-purpose sorter, improving on the current Daytona record holder by 65%.

Computer Science↗

Asynchronous GPU-based DEM solver embedded in commercial CFD software with polyhedral mesh support

A novel graphical processing unit-based discrete element method solver is introduced to improve stability, performance, and provide seamless integration into commercial or open-source computational fluid dynamics software. A key innovation is eliminating a need for network communication between solvers, which was previously required for cross-platform coupling. This is accomplished by a direct coupling method that employs dynamic-linked libraries. Furthermore, the solver optimizes memory usage by streamlining the particle-cell search algorithm by eliminating the cells' searching grid. This ensures the solver is compatible with a wide range of mesh types, providing high geometric flexibility. The approach simplifies the simulation process by directly incorporating computational fluid dynamics mesh information into the discrete element method solver. The performance analysis indicates about sixteen times boost in computational speed compared to benchmark central processing unit-based solvers. Finally, the solver's compatibility with polyhedral meshes, a vital advantage for complex geometries, is tested against a referenced study regarding the simulation of an immersed-tube fluidized bed.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Online State Estimation for Time-Varying Systems

The paper investigates the problem of estimating the state of a time-varying system with a linear measurement model; in particular, the paper considers the case where the number of measurements available can be smaller than the number of states. In lieu of a batch linear least-squares (LS) approach well-suited for static networks, where a sufficient number of measurements could be collected to obtain a full-rank design matrix the paper proposes an online algorithm to estimate the possibly time-varying state by processing measurements as and when available. The design of the algorithm hinges on a generalized LS cost augmented with a proximal-point-type regularization. With the solution of the regularized LS problem available in closed-form, the online algorithm is written as a linear dynamical system where the state is updated based on the previous estimate and based on the new available measurements. Conditions under which the algorithmic steps are in fact a contractive mapping are shown, and bounds on the estimation error are derived for different noise models. Numerical simulations are provided to corroborate the analytical findings.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Distributed out-of-memory NMF on CPU/GPU architectures

We propose an efficient distributed out-of-memory implementation of the non-negative matrix factorization (NMF) algorithm for heterogeneous high-performance-computing systems. The proposed implementation is based on prior work on NMFk, which can perform automatic model selection and extract latent variables and patterns from data. In this work, we extend NMFk by adding support for dense and sparse matrix operation on multi-node, multi-GPU systems. The resulting algorithm is optimized for out-of-memory problems where the memory required to factorize a given matrix is greater than the available GPU memory. Memory complexity is reduced by batching/tiling strategies, and sparse and dense matrix operations are significantly accelerated with GPU cores (or tensor cores when available). Input/output latency associated with batch copies between host and device is hidden using CUDA streams to overlap data transfers and compute asynchronously, and latency associated with collective communications (both intra-node and inter-node) is reduced using optimized NVIDIA Collective Communication Library (NCCL) based communicators. Benchmark results show significant improvement, from 32X to 76x speedup, with the new implementation using GPUs over the CPU-based NMFk. Good weak scaling was demonstrated on up to 4096 multi-GPU cluster nodes with approximately 25,000 GPUs when decomposing a dense 340 Terabyte-size matrix and an 11 Exabyte-size sparse matrix of density 10 -6 .

97 MATHEMATICS AND COMPUTING↗

Scalable Pattern Matching in Metadata Graphs via Constraint Checking

Pattern matching is a fundamental tool for answering complex graph queries. Unfortunately, existing solutions have limited capabilities: They do not scale to process large graphs and/or support only a restricted set of search templates or usage scenarios. Moreover, the algorithms at the core of the existing techniques are not suitable for today’s graph processing infrastructures relying on horizontal scalability and shared-nothing clusters, as most of these algorithms are inherently sequential and difficult to parallelize. In this article we present an algorithmic pipeline that bases pattern matching on constraint checking. The key intuition is that each vertex and edge participating in a match has to meet a set of constraints implicitly specified by the search template. These constraints can be verified independently and typically are less expensive to compute than searching the full template. The pipeline we propose generates these constraints and iterates over them to eliminate all the vertices and edges that do not participate in any match, thus reducing the background graph to a subgraph that is the union of all template matches—the complete set of all vertices and edges that participate in at least one match. Additional analysis can be performed on this annotated, reduced graph, such as full match enumeration, match counting, or computing vertex/edge centrality. Furthermore, a vertex-centric formulation for constraint checking algorithms exists, and this makes it possible to harness existing high-performance, vertex-centric graph processing frameworks. This technique (i) enables highly scalable pattern matching in metadata (labeled) graphs; (ii) supports arbitrary patterns with 100% precision; (iii) enables tradeoffs between precision and time-to-solution, while always selects all vertices and edges that participate in matches, thus offering 100% recall; and (iv) supports a set of popular data analytics scenarios. We implement our approach on top of HavoqGT, an open-source asynchronous graph processing framework, and demonstrate its advantages through strong and weak scaling experiments on massive scale real-world (up to 257 billion edges) and synthetic (up to 4.4 trillion edges) labeled graphs, respectively, and at scales (1,024 nodes / 36,864 cores), orders of magnitude larger than used in the past for similar problems. This article serves two purposes: First, it synthesises the knowledge accumulated during a long-term project. Second, it presents new system features, usage scenarios, optimizations, and comparisons with related work that strengthen the confidence that pattern matching based on iterative pruning via constraint checking is an effective and scalable approach in practice. The new contributions include the following: (i) We demonstrate the ability of the constraint checking approach to efficiently support two additional search scenarios that often emerge in practice, interactive incremental search and exploratory search. (ii) We empirically compare our solution with two additional state-of-the-art systems, Arabsque and TriAD. (iii) We show the ability of our solution to accommodate a more diverse range of datasets with varying properties, e.g., scale, skewness, label distribution, and match frequency. (iv) We introduce or extend a number of system features (e.g., work aggregation, load balancing, and the ability to cap the generated traffic) and design optimizations and demonstrate their advantages with respect to improving performance and scalability. (v) We present bottleneck analysis and insights into artifacts that influence performance. (vi) We present a theoretical complexity argument that motivates the performance gains we observe.

97 MATHEMATICS AND COMPUTING↗

The parallelism motifs of genomic data analysis

Genomic datasets are growing dramatically as the cost of sequencing continues to decline and small sequencing devices become available. Enormous community databases store and share these data with the research community, but some of these genomic data analysis problems require large-scale computational platforms to meet both the memory and computational requirements. These applications differ from scientific simulations that dominate the workload on high-end parallel systems today and place different requirements on programming support, software libraries and parallel architectural design. For example, they involve irregular communication patterns such as asynchronous updates to shared data structures. We consider several problems in high-performance genomics analysis, including alignment, profiling, clustering and assembly for both single genomes and metagenomes. We identify some of the common computational patterns or ‘motifs’ that help inform parallelization strategies and compare our motifs to some of the established lists, arguing that at least two key patterns, sorting and hashing, are missing. This article is part of a discussion meeting issue ‘Numerical algorithms for high-performance computational science’.

59 BASIC BIOLOGICAL SCIENCES↗

High performance graphics processor based computed tomography reconstruction algorithms for nuclear and other large scale applications.

The goal of this work is to develop a fast computed tomography (CT) reconstruction algorithm based on graphics processing units (GPU) that achieves significant improvement over traditional central processing unit (CPU) based implementations. The main challenge in developing a CT algorithm that is capable of handling very large datasets is parallelizing the algorithm in such a way that data transfer does not hinder performance of the reconstruction algorithm. General Purpose Graphics Processing (GPGPU) is a new technology that the Science and Technology (S&T) community is starting to adopt in many fields where CPU-based computing is the norm. GPGPU programming requires a new approach to algorithm development that utilizes massively multi-threaded environments. Multi-threaded algorithms in general are difficult to optimize since performance bottlenecks occur that are non-existent in single-threaded algorithms such as memory latencies. If an efficient GPU-based CT reconstruction algorithm can be developed; computational times could be improved by a factor of 20. Additionally, cost benefits will be realized as commodity graphics hardware could potentially replace expensive supercomputers and high-end workstations. This project will take advantage of the CUDA programming environment and attempt to parallelize the task in such a way that multiple slices of the reconstruction volume are computed simultaneously. This work will also take advantage of the GPU memory by utilizing asynchronous memory transfers, GPU texture memory, and (when possible) pinned host memory so that the memory transfer bottleneck inherent to GPGPU is amortized. Additionally, this work will take advantage of GPU-specific hardware (i.e. fast texture memory, pixel-pipelines, hardware interpolators, and varying memory hierarchy) that will allow for additional performance improvements.

97 MATHEMATICS AND COMPUTING↗