Search NASA⌕ Search

SEARCH · Search NASA

Results for “Kernel methods”

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 685 records · Page 38

The Borg Cube Simulation: Cosmological Hydrodynamics with CRK-SPH

A challenging requirement posed by next-generation observations is a firm theoretical grasp of the impact of baryons on structure formation. Cosmological hydrodynamic simulations modeling gas physics are vital in this regard. A high degree of modeling flexibility exists in this space, making it important to explore a range of methods in order to gauge the accuracy of simulation predictions. We present results from the first cosmological simulation using Conservative Reproducing Kernel Smoothed Particle Hydrodynamics (CRK-SPH). We employ two simulations: one evolved purely under gravity, and the other with nonradiative hydrodynamics. Each contains 2 × 2304 3 cold dark matter plus baryon particles in an 800 h -1 Mpc box. We compare statistics to previous nonradiative simulations including power spectra, mass functions, baryon fractions, and concentration. We find self-similar radial profiles of gas temperature, entropy, and pressure and show that a simple analytic model recovers these results to better than 40% over two orders of magnitude in mass. We quantify the level of nonthermal pressure support in halos and demonstrate that hydrostatic mass estimates are biased low by 24% (10%) for halos of mass 10 15 (10 13 ) h -1 M⊙. Here, we compute angular power spectra for the thermal and kinematic Sunyaev–Zel’dovich effects and find good agreement with the low-ℓ Planck measurements. Finally, artificial scattering between particles of unequal mass is shown to have a large impact on the gravity-only run, and we highlight the importance of better understanding this issue in hydrodynamic applications. This is the first in a simulation campaign using CRK-SPH, with future work including subresolution gas treatments.

79 ASTRONOMY AND ASTROPHYSICS↗

Thermal scattering law of C 2 H 4 n : Integrating experimental data with DFT calculations

Improvements in determination of the thermal scattering law of moderator materials (measuring, calculating and validating) are important for accurate prediction of neutron thermalization in nuclear systems. In this paper a methodology for producing thermal scattering libraries from the experimental data for polyethylene C 2 H 4 n is discussed. Double differential scattering cross section (DDSCS) experiments were performed at the Spallation Neutron Source of Oak Ridge National Laboratory (SNS ORNL). New scattering kernel evaluations, based on phonon spectrum for C 2 H 4 n , are created using the NJOY2016 code. Two different methods were used: direct and indirect geometry neutron scattering at ARCS and SEQUOIA, and VISION instruments, respectively, where the phonon spectrum was derived from the dynamical structure factor S(Q, ω ) obtained from the measured DDSCS. In order to compare and validate the newly created library, the experimental setup was simulated using MCNP6.1. Compared with the current ENDF/B-VII.1, the resulting RPI C 2 H 4 n libraries improved both double differential scattering and total scattering cross sections. A set of criticality benchmarks containing C 2 H 4 n from HEU-MET-THERM resulted in an overall improved calculation of K eff , although the libraries should be tested against benchmarks more sensitive to C 2 H 4 n . The DFT + oClimax method is used and is shown to be most comprehensive method for analysis of moderator materials. Finally, the importance of DFT + oClimax method lies in the fact that it can be validated against all data measured at VISION, ARCS and SEQUOIA, and experimental total scattering cross section measurements.

74 ATOMIC AND MOLECULAR PHYSICS↗

GPU Accelerated Sparse Cholesky Factorization

The solution of sparse symmetric positive definite linear systems is an important computational kernel in large-scale scientific and engineering modeling and simulation. We will solve the linear systems using a direct method, in which a Cholesky factorization of the coefficient matrix is performed using a right-looking approach and the resulting triangular factors are used to compute the solution. Sparse Cholesky factorization is compute intensive. In this work we investigate techniques for reducing the factorization time in sparse Cholesky factorization by offloading some of the dense matrix operations on a GPU. We will describe the techniques we have considered. We achieved up to 4x speedup compared to the CPU-only version.

Karsavuran, M Ozan↗

Sparse Data Acquisition on Emerging Memory Architectures

Emerging memory devices, such as resistive crossbars, have the capacity to store large amounts of data in a single array. Acquiring the data stored in large-capacity crossbars in a sequential fashion can become a bottleneck. We present practical methods, based on sparse sampling, to quickly acquire sparse data stored on emerging memory devices that support the basic summation kernel, reducing the acquisition time from linear to sub-linear. Experimental results show that at least an order of magnitude improvement in acquisition time can be achieved when the data is sparse.

97 MATHEMATICS AND COMPUTING↗

Scalable Heterogeneous Execution of a Coupled-Cluster Model with Perturbative Triples

The CCSD(T) coupled-cluster model with perturbative triples is considered a gold standard for computational modeling of the correlated behavior of electrons in molecular systems. A fundamental constraint is the relatively small global-memory capacity in GPUs compared to the main-memory capacity on host nodes, necessitating relatively smaller tile sizes for high-dimensional tensor contractions in NWChem's GPU-accelerated implementation of the CCSD(T) method. A coordinated redesign is described to address this limitation and associated data movement overheads, including a novel fused GPU kernel for a set of tensor contractions, along with inter-node communication optimization and data caching. The new implementation of GPU-accelerated CCSD(T) improves overall performance by 3.4x. Finally, we discuss the trade-offs in using this fused algorithm on current and future supercomputing platforms.

Kim, Jinsung↗

A Multigrid Framework for SN Discretizations of the Boltzmann Transport Equation with Application to Spatially Unstructured Grids

The space-angle multiple/semi-coarsening multigrid methods of [3] for S n discretizations of the Boltzmann transport equation was developed assuming rediscretizations of the streaming collision operators on the coarser spatial grids. These rediscretized operators are directly used in the kernels of the spatial coarse-level scalar flux integral equations, which undergo the multigrid process. Since the computation is essentially performed on these integral equations, it appears that the multigrid procedure is applied directly to the finest level scalar flux integral equation, rather than to the finest level S n system. However, this procedure is actually applied implicitly to the finest level S n system, though it would be beneficial to directly relate the multiple/semi-coarsening methods to the full S n system: being able to relate the multigrid procedure directly to the full S n system will lead to a general framework that can help guide the development of multigrid algorithms for more general S n discretization settings, such as when spatial rediscretization of the integral equation kernel can not be applied. In this latter case, applying an operator-dependent Galerkin coarsening procedure to the integral operator is difficult, if not realizable, because the kernel is dense and often given by its action of numerically sweeping over all the streaming-collision operators. In this paper, we develop and analyze such a multigrid framework for the S n system. We further relate this framework to a multigrid scheme that directly coarsens the scalar flux integral equation using a Galerkin coarsening approach. For this approach, we show that an intricate operator-dependent Galerkin coarsening procedure may not be needed. In particular, this paper develops a Galerkin-principle framework for S n discretizations of the Boltzmann equation, of which the analysis shows that simple interpolation operators can be used to coarsen the integral equations operators arising from the space-angle multiple/semi-coarsening methods.

97 MATHEMATICS AND COMPUTING↗

Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)

The development of modern processors exhibits two trends that complicate the optimizations of modern software. The first is the increasing sensitivity of processors' throughput to irregularities in computation. With more processors produced through a massive integration of simple cores, future systems will increasingly favor regular data-level parallel computations, but deviate from the needs of applications with complex patterns. Some evidences are already shown on Graphic Processing Units (GPU): Irregular data accesses (e.g., indirect references A[D[i]]) and conditional branches are limiting many GPU applications' performance at a level an order of magnitude lower than the peak of GPU. The second hardware trend is the growing gap between memory bandwidth and the aggregate speed—that is, the sum of all cores' computing power—of a Chip Multiprocessor (CMP). Despite the capped growth of the peak CPU speed, the aggregate speed of a CMP keeps increasing as more cores get into a single chip. It is expected that by 2018, node concurrency in an exascale system will increase by hundreds of times, whereas, memory bandwidth will expand by only 10 to 20 times. Consequently, data movement and storage is expected to consume more than 70% of the total system power. Bridging this gap is difficult; the complexities of modern CMP memory hierarchy make it even harder: Data cache becomes shared among computing units, and the sharing is often non-uniform—whether two computing units share a cache depends on their proximity and the level of the cache. On the recent IBM Power7 architecture, for instance, four hardware contexts (or SMT threads) in a core share the entire memory hierarchy, all cores in one chip share an on-chip L3 cache, and cores across chips share L3 and main memory through off-chip connections. These two trends complicate the translation of computing power into performance, especially for a program with either intensive data accesses or complex patterns in data accesses or control flow paths. Unfortunately, both attributes present and will persist in a class of important applications. For instance, many scientific simulations deal with a large volume of data. And meanwhile, as most real-world processes are non-uniform and evolving (e.g., the evolution of a galaxy or the process of a drug injection), both the computations and data accesses of these programs tend to be irregular and dynamically changing. Currently, the lack of support to these applications on modern CMP severely limits their performance. On GPU, as our recent study shows and other studies echo, performance enhancement of a factor of integers is possible when memory accesses or control flows are streamlined for a set of GPU applications. On multicore CPU, our studies show that traditional locality enhancement, for being oblivious to the new features of multicore memory hierarchy, may even cause large slowdown to data-intensive dynamic applications. The severity of the issues is expected to worsen as the two hardware trends continue. Some recent studies try to match software with the trends, but in a limited scope or manner. For irregularities on GPU, most studies focus on irregularities analyzable through static analysis (e.g., data accesses in regular loops). Dynamic irregularities are harder to address because the needed analysis and transformations typically have to happen at run time. Some other research resorts to hardware extensions, an actual adoption of which is unclear for the entailed space cost and complexity. For data locality, recent years have seen some exploitations of the new memory hierarchy on multicore for performance, but most of them are on process or thread scheduling, rather than program transformations. Our recent study reveals that program-level transformations may magnify the scheduling benefits by a factor of seven, concluding that program-level transformation should play a central role for data locality enhancement on modern CMP. But research in this direction has been sparse, and most have focused on data layout or cache performance modeling, rather than program transformations to match with the new memory hierarchy features. Overall, it is still an open question how to bridge the gap between dynamic computations and the two prominent properties of modern processors. The goal of this project is to develop a set of techniques and software tools to enhance the matching between memory accesses in dynamic simulations and the prominent features of modern and future CMP, alleviating the memory performance issues for petascale and exascale computing. This report summarizes the discoveries and products produced throughout this project. It includes free launch, a new software approach to overcoming the shortcomings of both methods; coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution; algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Supporting dynamic parallelism is important for GPU to benefit a broad range of applications. There are currently two fundamental ways for programs to exploit dynamic parallelism on GPU: a software-based approach with software-managed worklists, and a hardware-based approach through dynamic subkernel launches. Neither is satisfactory. The former is complicated to program and is often subject to some load imbalance; the latter suffers large runtime overhead. In this work, we propose free launch, a new software approach to overcoming the shortcomings of both methods. It allows programmers to use subkernel launches to express dynamic parallelism. It employs a novel compiler-based code transformation named subkernel launch removal to replace the subkernel launches with the reuse of parent threads. Coupled with an adaptive task assignment mechanism, the transformation reassigns the tasks in the subkernels to the parent threads with a good load balance. The technique requires no hardware extensions, immediately deployable on existing GPUs. It keeps the programming convenience of the subkernel launch-based approach while avoiding its large runtime overhead. Meanwhile, its superior load balancing makes it outperform manual worklist-based techniques by 3X on average. The work was published at the 48th Annual IEEE/ACM International Symposium on Microarchitecture (Micro'2015). A Graphic Processing Unit (GPU) system is typically equipped with many types of memory (e.g., global, constant, texture, shared, cache). Data placement determines what data are placed on which type of memory, essential for GPU memory performance. Prior optimizations of data placement always require a single view of a data object on memory, which limits the optimization effectiveness. In this work, we propose coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution. We demonstrate that under certain conditions, the multiple views can remain incoherent while facilitating enhanced data placement. We present a theorem and some compiler support to ensure the soundness of the usage of coherence-free multiview. We further develop reference-discerning data placement, a new way to enhance data placements on GPU. It enables more flexible data placements by using coherence-free multiview to leverage the slack in coherence requirement of some GPU programs. Experiments on three types of GPU systems show that, with less than 200KB space cost, the new data placement technique can provide a 1.6X average (up to 4.27X) speedup. In addition, we have examined algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Computing distances among data points is an essential part of many important algorithms in data analytics, graph analysis, and other domains. In each of these domains, developers have spent significant manual effort optimizing algorithms, often through novel applications of the triangle equality, in order to minimize the number of distance computations in the algorithms. In this work, we observe that many algorithms across these domains can be generalized as an instance of a generic distance-related abstraction. Based on this abstraction, we derive seven principles for correctly applying the triangular inequality to optimize distance-related algorithms. Guided by the findings, we develop Triangular OPtimizer (TOP), the first software framework that is able to automatically produce optimized algorithms that either matches or outperforms manually designed algorithms for solving distance-related problems. TOP achieves up to 237x speedups and 2.5X on average. The work has been published at the 32nd International Conference on Machine Learning (ICML'15) and the 41st International Conference on Very Large Data Bases (VLDB'15).

42 ENGINEERING↗

GalaxyFlow: upsampling hydrodynamical simulations for realistic mock stellar catalogues

ABSTRACT Cosmological N-body simulations of galaxies operate at the level of ‘star particles’ with a mass resolution on the scale of thousands of solar masses. Turning these simulations into stellar mock catalogues requires ‘upsampling’ the star particles into individual stars following the same phase-space density. In this paper, we introduce two new upsampling methods. First, we describe GalaxyFlow, a sophisticated upsampling method that utilizes normalizing flows to both estimate the stellar phase-space density and sample from it. Secondly, we improve on existing upsamplers based on adaptive kernel density estimation (KDE), using maximum likelihood estimation to fine-tune the bandwidth for such algorithms in a way that improves both the density estimation accuracy and upsampling results. We demonstrate our upsampling techniques on a neighbourhood of the Solar location in two simulated galaxies: Auriga 6 and h277. Both yield smooth stellar distributions that closely resemble the stellar densities seen in the Gaia DR3 catalogue. Furthermore, we introduce a novel multimodel classifier test to compare the accuracy of different upsampling methods quantitatively. This test confirms that GalaxyFlow more accurately estimates the density of the underlying star particles than methods based on KDE, at the cost of being more computationally intensive.

Lim, Sung Hak (ORCID:0000000330981092)↗

Time-Dependent Cryospheric Longwave Surface Emissivity Feedback in the Community Earth System Model

Frozen and unfrozen surfaces exhibit different longwave surface emissivities with different spectral characteristics, and outgoing longwave radiation and cooling rates are reduced for unfrozen scenes relative to frozen ones. In this paper physically realistic modeling of spectrally resolved surface emissivity throughout the coupled model components of the Community Earth System Model (CESM) is advanced, and implications for model high-latitude biases and feedbacks are evaluated. It is shown that despite a surface emissivity feedback amplitude that is, at most, a few percent of the surface albedo feedback amplitude, the inclusion of realistic, harmonized longwave, spectrally resolved emissivity information in CESM1.2.2 reduces wintertime Arctic surface temperature biases from -7.2 ± 0.9 K to -1.1 ± 1.2 K, relative to observations. The bias reduction is most pronounced in the Arctic Ocean, a region for which Coupled Model Intercomparison Project version 5 (CMIP5) models exhibit the largest mean wintertime cold bias, suggesting that persistent polar temperature biases can be lessened by including this physically based process across model components. The ice emissivity feedback of CESM1.2.2 is evaluated under a warming scenario with a kernel-based approach, and it is found that emissivity radiative kernels exhibit water vapor and cloud cover dependence, thereby varying spatially and decreasing in magnitude over the course of the scenario from secular changes in atmospheric thermodynamics and cloud patterns. Finally, accounting for the temporally varying radiative responses can yield diagnosed feedbacks that differ in sign from those obtained from conventional climatological feedback analysis methods.

54 ENVIRONMENTAL SCIENCES↗

Head-end process for the reprocessing of HTGR spent fuel

The reprocessing of HTGR spent fuels is in favor of the sustainable development of nuclear energy to realize the maximal use of nuclear resource and the minimum disposal of nuclear waste. The head-end of HTGR spent fuels reprocessing is different from that of the LWR spent fuels reprocessing because of the difference of spent fuel structure. The dismantling of the graphite spent fuel element and the highly effective dissolution of fuel kernel is the most difficult process in the head end of the reprocessing. Recently, some work on the head-end has been done in China. First, the electrochemical method with nitrate salt as electrolyte was studied to disintegrate the graphite matrix from HTGR fuel elements and release the coated fuel particles, to provide an option for the head-end technology of reprocessing. The results show that the graphite matrix can be effectively separated from the coated particle without any damage to the SiC layer. Secondly, the microwave-assisted heating was applied to dissolve the UO{sub 2} kernel from the crashed coated fuel particles. The ceramic UO{sub 2} as the solute has a good ability to absorb the microwave energy. The results of UO{sub 2} kernel dissolution from crushed coated particles by microwave heating show that the total dissolution percentage of UO{sub 2} is more than 99.99% after 3 times cross-flow dissolution with the following parameters: 8 mol/L HNO{sub 3}, temperature 100 Celsius degrees, initial ratio of solid to liquid 1.2 g/ml. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

NN interaction from the lattice

A method to calculate nuclear potentials by lattice QCD is reviewed. It utilizes the asymptotic long distances behavior of Nambu-Bethe-Salpeter (NBS) wave functions, which contain the information of scattering phase of QCD. Nuclear potential is defined as an energy-independent interaction kernel, so that the Schroedinger equation can reproduce NBS wave functions below the inelastic threshold. The resulting potential becomes faithful to the scattering phase because of the asymptotic behavior of the NBS wave functions. After the derivative expansion, the method is applied to the central and the tensor potentials. Their energy-independence is examined.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Performance Analysis and Optimal Node-aware Communication for Enlarged Conjugate Gradient Methods

Krylov methods are a key way of solving large sparse linear systems of equations but suffer from poor strong scalability on distributed memory machines. Furthermore, this is due to high synchronization costs from large numbers of collective communication calls alongside a low computational workload. Enlarged Krylov methods address this issue by decreasing the total iterations to convergence, an artifact of splitting the initial residual and resulting in operations on block vectors. In this article, we present a performance study of an enlarged Krylov method, Enlarged Conjugate Gradients (ECG), noting the impact of block vectors on parallel performance at scale. Most notably, we observe the increased overhead of point-to-point communication as a result of denser messages in the sparse matrix-block vector multiplication kernel. Additionally, we present models to analyze expected performance of ECG, as well as motivate design decisions. Most importantly, we introduce a new point-to-point communication approach based on node-aware communication techniques that increases efficiency of the method at scale.

97 MATHEMATICS AND COMPUTING↗

On the Kernel of a Two-Point Problem for a Partial Differential Equation of the Second Order in Time

We study the problem for a homogeneous partial differential equation of the second order with respect to time with given homogeneous two-point conditions in this variable and, in general, of the infinite order in the other (space) variable. It is proved that the analyzed problem possesses solely the trivial solution if the characteristic determinant is not identically equal to zero. In the case where the set of zeros of the characteristic determinant of this problem is nonempty, we propose a method for the construction of nontrivial solutions of the problem.

97 MATHEMATICS AND COMPUTING↗

From the Weyl quantization of a particle on the circle to number–phase Wigner functions

A generalized Weyl quantization formalism for a particle on the circle is shown to supply an effective method for defining the number–phase Wigner function in quantum optics. A Wigner function for the state ϱ{sup ^} and the kernel K for a particle on the circle is defined and its properties are analysed. Then it is shown how this Wigner function can be easily modified to give the number–phase Wigner function in quantum optics. Some examples of such number–phase Wigner functions are considered.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Large-scale GW calculations on pre-exascale HPC systems

The ab initio GW approach is a rigorous Green’s-function-based framework that can be employed to compute electronic excitation properties of a wide variety of materials such as extended systems, molecules, as well as confined and nanostructured materials with a very satisfactory accuracy. However, GW calculations on complex systems are often hindered by the high computational cost associated with the method. Here, in this study, we demonstrate how to significantly speedup GW calculations with a novel algorithm for the computationally intense kernel based on a non-blocking chunked cyclic communication scheme that minimizes latency in MPI messages, allows for overlapping of communication and computation, and improves cache usage in matrix multiplication operations. The optimized version of the code, implemented in the BerkeleyGW software package, is capable of scaling well to the full Cori computer at NERSC (Cray XC40) and achieves over 11 Peta FLOP/s of sustained performance. We showcase our work by performing large-scale GW calculations of defect structures on silicon, which require simulation cells containing over 1700 atoms, and which can now be efficiently executed in just a few minutes on large pre-exascale high-performance computing systems.

97 MATHEMATICS AND COMPUTING↗

Behavior-based approach to misuse detection of a simulated SCADA system - 339

This paper presents the initial findings in applying a behavior-based approach for detection of unauthorized activities in a simulated Supervisory Control and Data Acquisition (SCADA) system. Misuse detection of this type utilizes fault-free system telemetry to develop empirical models that learn normal system behavior. Future monitored telemetry sources that show statistically significant deviations from this learned behavior may indicate an attack or other unwanted actions. The experimental test bed consists of a set of Linux based enterprise servers that were isolated from a larger university research cluster. All servers are connected to a private network and simulate several components and tasks seen in a typical SCADA system. Telemetry sources included kernel statistics, resource usages and internal system hardware measurements. For this study, the Auto Associative Kernel Regression (AAKR) and Auto Associative Multivariate State Estimation Technique (AAMSET) are employed to develop empirical models. Prognostic efficacy of these methods for computer security used several groups of signals taken from available telemetry classes. The Sequential Probability Ratio Test (SPRT) is used along with these models for intrusion detection purposes. The different intrusion types shown include host/network discovery, DoS, brute force login, privilege escalation and malicious exfiltration actions. For this study, all intrusion types tested displayed alterations in the residuals of much of the monitored telemetry and were able to be detected in all signal groups used by both model types. The methods presented can be extended and implemented to industries besides nuclear that use SCADA or business-critical networks. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A massively parallel and scalable multi-CPU material point method

Harnessing the power of modern multi-GPU architectures, we present a massively parallel simulation system based on the Material Point Method (MPM) for simulating physical behaviors of materials undergoing complex topological changes, self-collision, and large deformations. Our system makes three critical contributions. First, we introduce a new particle data structure that promotes coalesced memory access patterns on the GPU and eliminates the need for complex atomic operations on the memory hierarchy when writing particle data to the grid. Second, we propose a kernel fusion approach using a new Grid-to-Particles-to-Grid (G2P2G) scheme, which efficiently reduces GPU kernel launches, improves latency, and significantly reduces the amount of global memory needed to store particle data. Finally, we introduce optimized algorithmic designs that allow for efficient sparse grids in a shared memory context, enabling us to best utilize modern multi-GPU computational platforms for hybrid Lagrangian-Eulerian computational patterns. We demonstrate the effectiveness of our method with extensive benchmarks, evaluations, and dynamic simulations with elastoplasticity, granular media, and fluid dynamics. In comparisons against an open-source and heavily optimized CPU-based MPM codebase [Fang et al. 2019] on an elastic sphere colliding scene with particle counts ranging from 5 to 40 million, our GPU MPM achieves over 100x per-time-step speedup on a workstation with an Intel 8086K CPU and a single Quadro P6000 GPU, exposing exciting possibilities for future MPM simulations in computer graphics and computational science. Moreover, compared to the state-of-the-art GPU MPM method [Hu et al. 2019a], we not only achieve 2x acceleration on a single GPU but our kernel fusion strategy and Array-of-Structs-of-Array (AoSoA) data structure design also generalizes to multi-GPU systems. Our multi-GPU MPM exhibits near-perfect weak and strong scaling with 4 GPUs, enabling performant and large-scale simulations on a 10243 grid with close to 100 million particles with less than 4 minutes per frame on a single 4-GPU workstation and 134 million particles with less than 1 minute per frame on an 8-GPU workstation.

Wang, Xinlei↗

Quantitative Performance Assessment of Proxy Apps and Parents (Report for ECP Proxy App Project Milestone ADCD-504-28)

The ECP Proxy Application Project has an annual milestone to assess the state of ECP proxy applications and their role in the overall ECP ecosystem. Our FY22 March/April milestone (ADCD- 504-28) proposed to: Assess the fidelity of proxy applications compared to their respective parents in terms of kernel and I/O behavior, and predictability. Similarity techniques will be applied for quantitative comparison of proxy/parent kernel behavior. MACSio evaluation will continue and support for OpenPMD backends will be explored. The execution time predictability of proxy apps with respect to their parents will be explored through a carefully designed scaling study and code comparisons. Note that in this FY, we also have quantitative assessment milestones that are due in September and are, therefore, not included in the description above or in this report. Another report on these deliverables will be generated and submitted upon completion of these milestones. To satisfy this milestone, the following specific tasks were completed: Study the ability of MACSio to represent I/O workloads of adaptive mesh codes. Re-define the performance counter groups for contemporary Intel and IBM platforms to better match specific hardware components and to better align across platforms (make cross-platform comparison more accurate). Perform cosine similarity study based on the new performance counter groups on the Intel and IBM P9 platforms. Perform detailed analysis of performance counter data to accurately average and align the data to maintain phases across all executions and develop methods to reduce the set of collected performance counters used in cosine similarity analysis. Apply a quantitative similarity comparison between proxy and parent CPU kernels. Perform scaling studies to understand the accuracy of predictability of the parent performance using its respective proxy application. This report presents highlights of these efforts.

97 MATHEMATICS AND COMPUTING↗