Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel Optimization 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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

167-PFlops deep learning for electron microscopy: from learning physics to atomic manipulation

An artificial intelligence system called MENNDL, which used 25,200 NVIDIA Volta GPUs on Oak Ridge National Laboratory's Summit machine, automatically designed an optimal deep learning network in order to extract structural information from raw atomic-resolution microscopy data. In a few hours, MENNDL creates and evaluates millions of networks using a scalable, parallel, asynchronous genetic algorithm augmented with a support vector machine to automatically find a superior deep learning network topology and hyper-parameter set than a human expert can find in months. For the application of electron microscopy, the system furthers the goal of improving our understanding of the electron-beam-matter interactions and real-time image-based feedback, which enables a huge step beyond human capacity towards nanofabricating materials automatically. MENNDL has been scaled to the 4,200 available nodes of Summit achieving a measured 152.5 PFlops, with an estimated sustained performance of 167 PFlops when the entire machine is available.

Patton, Robert↗
DOE OSTI2025

A Scalable Interior‐Point Gauss–Newton Method for PDE‐Constrained Optimization With Bound Constraints

Here, we present a scalable approach to solve a class of partial differential equation (PDE)‐constrained optimization problems with bound constraints. This approach utilizes a robust full‐space interior‐point (IP)‐Gauss–Newton optimization method. To cope with the poorly‐conditioned IP‐Gauss–Newton saddle‐point linear systems that need to be solved approximately, once per optimization step, we propose two spectrally related preconditioners. These preconditioners leverage the limited informativeness of data in regularized PDE‐constrained optimization problems. A block Gauss–Seidel preconditioner is proposed for the GMRES‐based solution of the IP‐Gauss–Newton linear systems. It is shown, for a large‐class of PDE‐ and bound‐constrained optimization problems, that the spectrum of the block Gauss–Seidel preconditioned IP‐Gauss–Newton matrix is asymptotically independent of discretization and is not impacted by the ill‐conditioning that notoriously plagues interior‐point methods. We exploit symmetry of the IP‐Gauss–Newton linear systems and propose a regularization and log‐barrier Hessian preconditioner for the preconditioned conjugate gradient (PCG)‐based solution of the equivalent IP‐Gauss–Newton–Schur complement linear systems. The eigenvalues of the block Gauss–Seidel preconditioned IP‐Gauss–Newton matrix, that are not equal to one, are identical to the eigenvalues of the regularization and log‐barrier Hessian preconditioned Schur complement matrix. The scalability of the approach is demonstrated on two example problems. The numerical solution of these optimization problems is shown to require a discretization independent number of IP‐Gauss–Newton linear solves. Furthermore, the linear systems are solved in a discretization and IP ill‐conditioning independent number of preconditioned Krylov subspace iterations. The parallel scalability of the preconditioner, achieved via algebraic multigrid component solvers when applicable, and the aforementioned algorithmic scalability permits a parallel scalable means to compute solutions of a large class of PDE‐ and bound‐constrained problems.

PDE-constrained optimization↗
DOE OSTI2022

Receptance coupling substructure analysis and chatter frequency-informed machine learning for milling stability

This paper describes a milling stability identification approach that simultaneously considers: physics-based models for the tool tip frequency response functions and stability predictions; the binary result from a milling test (automatically labeled as stable or unstable based on frequency content); chatter frequency when an unstable result is obtained; and user risk tolerance. The algorithm applies probabilistic Bayesian machine learning with adaptive, parallelized Markov Chain Monte Carlo sampling to update the probability of stability with each milling test. Furthermore, the result is a robust solution for rapid convergence to optimized milling parameters for maximum metal removal rate using all available information.

42 ENGINEERING↗
DOE OSTI2025

Track reconstruction as a service for collider physics

Optimizing charged-particle track reconstruction algorithms is crucial for efficient event reconstruction in Large Hadron Collider (LHC) experiments due to their significant computational demands. Existing track reconstruction algorithms have been adapted to run on massively parallel coprocessors, such as graphics processing units (GPUs), to reduce processing time. Nevertheless, challenges remain in fully harnessing the computational capacity of coprocessors in a scalable and non-disruptive manner. This paper proposes an inference-as-a-service approach for particle tracking in high energy physics experiments. To evaluate the efficacy of this approach, two distinct tracking algorithms are tested: Patatrack, a rule-based algorithm, and Exa.TrkX, a machine learning-based algorithm. The as-a-service implementations show enhanced GPU utilization and can process requests from multiple CPU cores concurrently without increasing per-request latency. The impact of data transfer is minimal and insignificant compared to running on local coprocessors. This approach greatly improves the computational efficiency of charged particle tracking, providing a solution to the computing challenges anticipated in the High-Luminosity LHC era.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Implementation of a Mesh refinement algorithm into the quasi-static PIC code QuickPIC

Plasma-based acceleration (PBA) has emerged as a promising candidate for the accelerator technology used to build a future linear collider and/or an advanced light source. In PBA, a trailing or witness particle beam is accelerated in the plasma wave wakefield (WF) created by a laser or particle beam driver. The WF is often nonlinear and involves the crossing of plasma particle trajectories in real space and thus particle-in-cell methods are used. The distance over which the drive beam evolves is several orders of magnitude larger than the wake wavelength. This large disparity in length scales is amenable to the quasi-static approach. Three-dimensional (3D), quasi-static (QS), particle-in-cell (PIC) codes, e.g., QuickPIC, have been shown to provide high fidelity simulation capability with 2-4 orders of magnitude speedup over 3D fully explicit PIC codes. In PBA, the witness beam needs to be matched to the focusing forces of the WF to reduce the emittance growth. In some linear collider designs, the matched spot size of the witness beam can be 2 to 3 orders of magnitude smaller than the spot size (and wavelength) of the wakefield. Such an additional disparity in length scales is ideal for mesh refinement where the WF within the witness beam is described on a finer mesh than the rest of the WF. A mesh refinement scheme is described that has been implemented into the 3D QS PIC code, QuickPIC. Very fine (high) resolution is used in a small spatial region that includes the witness beam and progressively coarser resolutions in the rest of the simulation domain. A fast multigrid Poisson solver has been implemented for the field solve on the refined meshes and a Fast Fourier Transform (FFT) based Poisson solver is used for the coarse mesh. The code has been parallelized with both MPI and OpenMP, and the parallel scalability has also been improved by using pipelining. A preliminary adaptive mesh refinement technique is described to optimize the computational time for simulations with an evolving witness beam size. Several test problems are used to verify that the mesh refinement algorithm provides accurate results. Additionally, the results are benchmarked against highly resolved simulations exhibiting near-azimuthal symmetry, performed using QPAD—a novel hybrid QS PIC code that uses a PIC description in the coordinates (r, ct – z) and a gridless description in the azimuthal angle, Φ.

Linear collider↗
DOE OSTI2020

Virtual-Voltage Partition-Based Approach to Optimal Transmission Switching

This article deals with optimal transmission switching (OTS) problems involving binary decisions about network topology and nonconvex power flow constraints. We adopt a semidefinite programming formulation for the optimal power flow (OPF) problem that, however, remains nonconvex due to the presence of discrete variables and bilinear products between the decision variables. To tackle the latter, we introduce a physically inspired, virtual-voltage approximation that leads to provable lower and upper bounds on the solution of the original problem. To deal with the exponential complexity caused by the discrete variables, we introduce a graph partition-based algorithm that breaks the problem into several parallel mixed-integer subproblems of smaller size. Simulations on the IEEE bus test cases demonstrate the high degree of accuracy and affordable computational requirements of our approach.

42 ENGINEERING↗
DOE OSTI2025

Computational Algorithms for Unit Commitment with AC Power Flows (Final Report)

Security-constrained unit commitment (SCUC) is a key component in power system operations. When AC power flow constraints are considered in the SCUC model (AC-SCUC), the problem becomes extremely difficult due to its discrete and non-convex nature, as described in “Grid Optimization Competition Challenge 3 Problem Formulation (GOCC)”. There are four main challenges: (i) Discrete decisions regarding unit online/offline status and start-up/shut-down procedures for every single unit. The number of discrete decision variables increases considerably when a system integrates multiple generators; (ii) Configuration-based combined-cycle formulations, and multi-commodity models that include ramping products, spin/non-spin products, and regulation up/down products. The combined-cycle units introduce additional discrete decision variables and auxiliary service products further complicate the model by connecting multi-commodity products’ continuous and discrete variables; (iii) SCUC models with AC power flow constraints are far more complex due to massive bilinear terms in the large-scale nonlinear power balance equations. The nonlinear power balance equations are further complicated by the discrete step control variables of shunts; (iv) N − 1 contingency analysis. The size of the model increases linearly with the number of contingencies considered, greatly increasing the size of the optimization model. Accordingly, there is an emergent need to develop a robust algorithm capable of deriving a high-quality solution in a short time and passing through contingency tests simultaneously. In this project, we explore innovative techniques to address this challenging problem by integrating advanced polyhedral theory, approximation methods, relaxation strategies, decomposition techniques, and parallel computing. Each technique approaches the problem from a different perspective, leveraging its specific strengths to tackle distinct challenges. Each individual method has demonstrated its effectiveness in the PI’s previous research. Their integration is expected to significantly reduce the computational time required to solve the proposed complex problem. Successful completion of this project has the potential to transform the industry by enhancing optimization solvers capable of handling large-scale day-ahead energy market clearing models within strict time constraints, while incorporating AC power flow constraints. This advancement will lead to reduced overall generation costs and, consequently, increased social welfare.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2020

Parallel performance of algebraic multigrid domain decomposition

Algebraic multigrid (AMG) is a widely used scalable solver and preconditioner for large-scale linear systems resulting from the discretization of a wide class of elliptic PDEs. While AMG has optimal computational complexity, the cost of communication has become a significant bottleneck that limits its scalability as processor counts continue to grow on modern machines. This article examines the design, implementation, and parallel performance of a novel algorithm, algebraic multigrid domain decomposition (AMG-DD), designed specifically to limit communication. The goal of AMG-DD is to provide a low-communication alternative to standard AMG V-cycles by trading some additional computational overhead for a significant reduction in communication cost. Numerical results show that AMG-DD achieves superior accuracy per communication cost compared with AMG, and speedup over AMG is demonstrated on a large GPU cluster.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Dose–volume histogram evaluation scheme for beam orientations in intensity-modulated radiation therapy

Beam orientations in intensity-modulated radiation therapy (IMRT) is an important, but difficult NP-hard optimization problem. A considerable part of this difficulty is due to the essential prerequisite of defining, appropriately, counter-intuitive criteria for a mathematical plan evaluation that coincides with the clinical judgment of the considered plan. Moreover, the quality of beam directions depends heavily on their intensities. Usually a stochastic selector is utilized for optimizing beam orientations, and the intensities are thus calculated many thousands of times, each time for a different selection of beam directions, resulting in excessive time complexity. Therefore, selecting an appropriate set of beam directions in IMRT is still a time-consuming manual trial-and-error search procedure that depends on intuition and empirical knowledge. To overcome these difficulties, this work utilizes the concept of dose–volume histogram (DVH), which is one of the main recognized quantitative measurement tools used for plan judgment, and presents a DVH evaluation scheme that parallelizes plan evaluation with clinical plan judgment. This scheme is then utilized to optimize the intensities of each individual beam, enabling it to be represented by its DVHs, produced for the different irradiated volumes. Finally, an evolutionary algorithm is employed to search for the best combination of those DVHs reflecting the clinical optimal. The presented methods have been applied to three real clinical examples. In comparison to the standard equally spaced beam plans, improvements were reported in all three cases. In comparison with the greedy algorithm, improvements were reported in those examples requiring more than five beam directions.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2025

A scoping study of far-SOL main-wall protection limiters for steady-state operation of compact pilot plant tokamaks

We present a novel method for handling steady-state heat fluxes incident on the main wall of pilot plant-scale magnetic fusion devices, based on the utilization of protection limiters in the far scrape-off layer (SOL). This method helps avoid large plasma-wall gaps, without excessively compromising blanket performance. We present an optimization algorithm for determining the appropriate size and scale of these protection limiters given (1) probability distributions of SOL plasma parameters and (2) assumed risk tolerance. As part of this optimization, we have developed an analytic description of parallel heat fluxes across limiter shadows, and an objective cost function (the ‘Far-SOL Marginal Cost’) to quantify the impact that different main-wall thermal management design choices have on reactor capital cost. Applying the model to a midscale fusion pilot plant concept shows that making use of far-SOL protection limiters can reduce capital costs on the order of $500 M, relative to naively increasing the plasma-wall gap. Our analysis demonstrates that the far-SOL power decay length is the highest-leverage plasma assumption for thermal loading of the first wall, and the primary cost driver for main wall thermal management. The relative cost efficiency of protection limiters increases as assumptions on the far-SOL heat flux become more pessimistic. The concepts described in this paper motivate the further development of far-SOL protection limiters as part of larger efforts to design economical core-edge-wall compatible solutions for a fusion pilot plant.

Design under uncertainty↗
DOE OSTI2022

Algorithm 1028: VTMOP: Solver for Blackbox Multiobjective Optimization Problems

VTMOP is a Fortran 2008 software package containing two Fortran modules for solving computationally expensive bound-constrained blackbox multiobjective optimization problems. VTMOP implements the algorithm of [32], which handles two or more objectives, does not require any derivatives, and produces well-distributed points over the Pareto front. The first module contains a general framework for solving multiobjective optimization problems by combining response surface methodology, trust region methodology, and an adaptive weighting scheme. The second module features a driver subroutine that implements this framework when the objective functions can be wrapped as a Fortran subroutine. Lastly, support is provided for both serial and parallel execution paradigms, and VTMOP is demonstrated on several test problems as well as one real-world problem in the area of particle accelerator optimization.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Final Technical Report: Mathematical Foundations for Uncertainty Quantification in Materials Design

We developed path-wise information theory-based and goal-oriented sensitivity analysis and parameter identification methods for complex high-dimensional dynamics and in particular of non-equilibrium extended molecular systems. The combination of these novel methodologies provided the first methods in the literature which are capable to handle UQ questions for stochastic complex systems with some or all of the following features: (a) multi-scale stochastic models such as (bio)chemical reaction networks, with a very large number of parameters, (b) spatially distributed systems such as Kinetic Monte Carlo or Langevin Dynamics, (c) non-equilibrium processes typically associated with coupled physico-chemical mechanisms, driven boundary conditions, hybrid micro-macro systems, etc. A particular computational challenge arises in simulations of multi-scale reaction networks and molecular systems. Mathematical techniques were applied to in silico prediction of novel materials with emphasis on the effect of microstructure on model uncertainty quantification (UQ). We outline acceleration methods to make calculations of real chemistry feasible followed by two complementary tasks on structure optimization and microstructure-induced UQ.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Optimizing High Performance Markov Clustering for Pre-Exascale Architectures

HipMCL is a high-performance distributed memory implementation of the popular Markov Cluster Algorithm (MCL) and can cluster large-scale networks within hours using a few thousand CPU-equipped nodes. It relies on sparse matrix computations and heavily makes use of the sparse matrix-sparse matrix multiplication kernel (SpGEMM). The existing parallel algorithms in HipMCL are not scalable to Exascale architectures, both due to their communication costs dominating the runtime at large concurrencies and also due to their inability to take advantage of accelerators that are increasingly popular. In this work, we systematically remove scalability and performance bottlenecks of HipMCL. We enable GPUs by performing the expensive expansion phase of the MCL algorithm on GPU. Additionally, we propose a CPU-GPU joint distributed SpGEMM algorithm called pipelined Sparse SUMMA and integrate a probabilistic memory requirement estimator that is fast and accurate. Furthermore, we develop a new merging algorithm for the incremental processing of partial results produced by the GPUs, which improves the overlap efficiency and the peak memory usage. We also integrate a recent and faster algorithm for performing SpGEMM on CPUs. We validate our new algorithms and optimizations with extensive evaluations. With the enabling of the GPUs and integration of new algorithms, HipMCL is up to 12.4x faster, being able to cluster a network with 70 million proteins and 68 billion connections just under 15 minutes using 1024 nodes of ORNL's Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Multigrid treatment of implicit continuum diffusion

Implicit treatment of diffusive terms of various differential orders common in continuum mechanics modeling, such as computational fluid dynamics, is investigated with spectral and multigrid algorithms in non-periodic 2D domains. In doubly periodic time dependent problems these terms are handled efficiently by spectral methods, but in non-periodic systems solved with distributed memory parallel computing and 2D domain decomposition, this efficiency is lost for a large number of processors. Here, we built and present here a multigrid algorithm for these types of problems that outperforms a spectral solution employing the highly optimized FFTW library. This solver is suitable for high performance computing and may be able to efficiently treat implicit diffusion of arbitrary order by introducing auxiliary equations of lower order. We test these solvers for fourth and sixth order diffusion with harmonic test functions as well as turbulent 2D magnetohydrodynamic simulations. It is also shown that an anisotropic operator without mixed-derivative terms improves model accuracy and speed, and we examine the impact that the various diffusion operators have on the energy, the enstrophy, and the qualitative aspect of a simulation.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Distributed fiber sensing systems for 3D combustion temperature field monitoring in coal-fired boilers using optically generated acoustic waves (Final Report)

In this project, we have developed and tested three kinds of fiber optic sensing systems for real time monitoring of temperature variations within an industrial scale boiler furnace. The fiber optic sensing systems target spatial and temporal distributions of high temperature profiles in a boiler furnace in fossil power plants. The reconstructed temperature profile will provide critical input for the control mechanisms to optimize the combustion process. This temperature profile will address the essential problem for fossil power plants in achieving higher efficiency and fewer pollutant emissions. Acoustic pyrometer systems have been used to reconstruct temperature field of power plant boilers based on measuring TOF (times-of-flight) of sound waves along some straight paths in a 2D cross-section of the boiler. In this project, optically generated acoustic signals from a fiber optic sensing system have replaced the acoustic signals generated from an electrical transducer. A 3D reconstruction algorithm replaced the previous 2D model. In this project, three kinds of fiber optic sensing systems have been developed and tested. They are fiber optic sensing system I, fiber optic sensing system II (Distributed Sensing System I) and fiber optic sensing system III (Distributed Sensing System II). For fiber optic sensing system I, the fiber optic ultrasound generator acts as a signal generator. A microphone, hydrophone or other electronic devices serve as a signal receiver. In this system, there are one generator and one receiver. Distance test, water temperature test, air temperature test, air temperature reconstruction, and GE ISBF pilot test were performed by Fiber optic sensing system I. The fiber optic sensing system I successfully detected temperature in all these tests. We got 2D temperature reconstruction results by using the fiber optic sensing system I and it matched the reference data. The fiber optic sensing system I successfully survived in GE ISBF boiler environment (480 °F). For fiber optic sensing system II (Distributed Sensing System I), it is an all optical ultrasound system. The fiber optic ultrasound generator acts as a signal generator. Fiber Bragg Grating (FBG) and Fabry-Perot (FP) sensor act as a signal receiver. In this system, there is one generator and one receiver. Aluminum plate temperature test, furnace high temperature test, and GE ISBF pilot test were performed by the fiber optic sensing system II. Fiber optic sensing system II successfully detected the temperature in all these tests. The fiber optic sensing system II successfully survived at up to 700 °C furnace environment and 320 °C GE ISBF boiler environment. For fiber optic sensing system III (Distributed Sensing System II), it is also an all optical ultrasound system. The fiber optic ultrasound generator acts as a signal generator. Multiple FP fiber sensors act as signal receivers. In this system, there is one generator and three receivers. Three GE ISBF pilot tests were performed by fiber optic sensing system III. The fiber optic sensing system III survived in the cold flow tests in GE’s ISBF pilot test facility. However, we didn’t get high temperature data by using this system since the nanosecond laser issues. During the period of the project, test trials, data simulation and algorithm optimization was performed successfully. For real time temperature field construction, the sampling rate must be fast enough to capture the field variations. The technology of Code-division multiple access (CDMA) is well studied which could allow parallel multiplexing, even if signals overlap in time or frequencies. Moreover, it has been known that extending the length of signal significantly improves SNR. For acoustic signals, these multiplexing techniques have also been widely used, mainly for sonar and acoustic communications. The CDMA modulation technique has been proposed and studied to guarantee high network throughput, low channel access delay and low energy consumption. We have studied the temperature field reconstruction using Gaussian Radial Basis Functions (GRBF)-based approximation approach. Reconstruction of 3D temperature field using Neural Networks with measured TOF and known propagation paths is feasible. 2D and 3D temperature field reconstruction simulation results are achieved. The milestone status is shown in Table 1. We finished milestone 1-8 and milestone 10. For milestone 9, we did three pilot tests by using the fiber optic sensing system III (Distributed Sensing System II) at GE Power. However, due to the failure of the ns laser, we did not get the temperature results. We conducted some additional tasks that were not originally proposed: 1) We fabricated a fiber optic sensing system I and did a pilot test based on this system. 2) In the proposal, we proposed two pilot tests at GE Power. In reality, we finished at least seven pilot tests at GE Power. GE Power has made a lot of efforts for supporting the pilot tests. 3) We got a simulation results based on CDMA. In summary, most of the tasks have been accomplished. The outcome of this project removed a few barriers that hinder the achievement of the final product of the distributed sensing systems. With the successful accomplishment of this project, a prototype of the fiber optic sensing system can be fabricated to attract more interests from companies and other funding agencies.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

Parallel physics-informed neural networks via domain decomposition

Here we develop a distributed framework for the physics-informed neural networks (PINNs) based on two recent extensions, namely conservative PINNs (cPINNs) and extended PINNs (XPINNs), which employ domain decomposition in space and in time-space, respectively. This domain decomposition endows cPINNs and XPINNs with several advantages over the vanilla PINNs, such as parallelization capacity, large representation capacity, efficient hyperparameter tuning, and is particularly effective for multi-scale and multi-physics problems. Here, we present a parallel algorithm for cPINNs and XPINNs constructed with a hybrid programming model described by MPI + X, where X ∈ {CPUs, GPUs}. The main advantage of cPINN and XPINN over the more classical data and model parallel approaches is the flexibility of optimizing all hyperparameters of each neural network separately in each subdomain. We compare the performance of distributed cPINNs and XPINNs for various forward problems, using both weak and strong scalings. Our results indicate that for space domain decomposition, cPINNs are more efficient in terms of communication cost but XPINNs provide greater flexibility as they can also handle time-domain decomposition for any differential equations, and can deal with any arbitrarily shaped complex subdomains. To this end, we also present an application of the parallel XPINN method for solving an inverse diffusion problem with variable conductivity on the United States map, using ten regions as subdomains.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

CSRI Summer Proceedings 2020

The Computer Science Research Institute (CSRI) brings university faculty and students to Sandia for focused collaborative research on Department of Energy (DOE) computer and computational science problems. The institute provides an opportunity for university researchers to learn about problems in computer and computational science at DOE laboratories. Participants conduct leading-edge research, interact with scientists and engineers at the laboratories, and help transfer results of their research to programs at the labs. Some specific CSRI research interest areas are: scalable solvers, optimization, adaptivity and mesh refinement, graph-based, discrete, and combinatorial algorithms, uncertainty estimation, mesh generation, dynamic load-balancing, virus and other malicious-code defense, visualization, scalable cluster computers, data-intensive computing, environments for scalable computing, parallel input/output, advanced architectures, and theoretical computer science. The CSRI Summer Program is organized by CSRI and typically includes the organization of a weekly seminar series and the publication of a summer proceedings. In 2020, the CSRI summer program was executed completely virtually; all student interns worked from home, due to the COVID-19 pandemic.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Bit-GraphBLAS: Bit-Level Optimizations of Matrix-Centric Graph Processing on GPU

In the graph data structure like adjacency matrix, the connectivity of two nodes can be sufficiently represented using only 1 bit, but they are generally treated as 32-bit full-precision in state-of-the-art graph frameworks to adopt common sparse format such as CSR. Meanwhile, bit-level parallelism has recently be explored to have high-performance potential and low storage requirement on GPUs with dense bit-tiles. To fill the gap, our solution is a hierarchical storage format that contains the bit-indexing base and dense bit-tile units. Inherently, the granularity of the bit-tile is an essential factor in achieving both storage compression and GPU parallelism. How to find a sweet spot that trades off between avoiding sparsity and exploiting is comprehensively researched in this work. In the experiment, we evaluate the proposed storage format and algorithms on modern generation GPUs, including Pascal and Volta, to figure out critical software co-designs in conjunction with existing hardware-specific optimization.

Chen, Jou-An↗