Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “memory usage”

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 217 records · Page 12

Design and Implementation of Papyrus: Parallel Aggregate Persistent Storage

A surprising development in recently announced HPC platforms is the addition of, sometimes massive amounts of, persistent (nonvolatile) memory (NVM) in order to increase memory capacity and compensate for plateauing I/O capabilities. However, there are no portable and scalable programming interfaces using aggregate NVM effectively. This paper introduces Papyrus: a new software system built to exploit emerging capability of NVM in HPC architectures. Papyrus (or Parallel Aggregate Persistent -YRU- Storage) is a novel programming system that provides features for scalable, aggregate, persistent memory in an extreme-scale system for typical HPC usage scenarios. Papyrus mainly consists of Papyrus Virtual File System (VFS) and Papyrus Template Container Library (TCL). Papyrus VFS provides a uniform aggregate NVM storage image across diverse NVM architectures. It enables Papyrus TCL to provide a portable and scalable high-level container programming interface whose data elements are distributed across multiple NVM nodes without requiring the user to handle complex communication, synchronization, replication, and consistency model. We evaluate Papyrus on two HPC systems, including UTK Beacon and NERSC Cori, using real NVM storage devices.

Kim, Jungwon↗

Efficient Checkpointing of Virtual Machines using Virtual Machine Introspection

Cloud Computing environments rely heavily on system-level virtualization. This is due to the inherent benefits of virtualization including fault tolerance through checkpoint/restart (C/R) mechanisms. Because clouds are the abstraction of large data centers and large data centers have a higher potential for failure, it is imperative that a C/R mechanism for such an environment provide minimal latency as well as a small checkpoint file size. Recently, there has been much research into C/R with respect to virtual machines (VM) providing excellent solutions to reduce either checkpoint latency or checkpoint file size. However, these approaches do not provide both. This paper presents a method of checkpointing VMs by utilizing virtual machine introspection (VMI). Through the usage of VMI, we are able to determine which pages of memory within the guest are used or free and are better able to reduce the amount of pages written to disk during a checkpoint. We have validated this work by using various benchmarks to measure the latency along with the checkpoint size. With respect to checkpoint file size, our approach results in file sizes within 24% or less of the actual used memory within the guest. Additionally, the checkpoint latency of our approach is up to 52% faster than KVM s default method.

Aderholdt, Ferrol↗

Critical Points Based Register-Concurrency Autotuning for GPUs

The unprecedented prevalence of GPGPU is largely attributed to its abundant on-chip register resources, which allow massively concurrent threads and extremely fast context switch. However, due to internal memory capacity constraints, there is a tradeoff between the per-thread register usage and the overall concurrency. This becomes a design problem in terms of performance tuning, since the performance “sweet spot” which can be significantly affected by these two factors is generally unknown beforehand. In this paper, we propose an effective autotuning solution to quickly and efficiently select the optimal number of registers perthread for delivering the best GPU performance. Experiments on three generations of GPUs (Nvidia Fermi, Kepler and Maxwell) demonstrate that our simple strategy can achieve an average of 10% performance improvement while a max of 50% over the original version without modifying the user program. Additionally, to reduce local cache misses due to register spilling and further improve performance, we explore three optimization schemes (i.e. bypass L1 for global memory access, enlarge local L1 cache and spill into shared memory) and discuss their impact on performance on a Kepler GPU.

auto-tuning↗

High Performance Data Transfer for Distributed Data Intensive Sciences

We report on the development of ZX software providing high performance data transfer and encryption. The design scales in: computation power, network interfaces, and IOPS while carefully balancing the available resources. Two U.S. patent-pending algorithms help tackle data sets containing lots of small files and very large files, and provide insensitivity to network latency. It has a cluster-oriented architecture, using peer-to-peer technologies to ease deployment, operation, usage, and resource discovery. Its unique optimizations enable effective use of flash memory. Using a pair of existing data transfer nodes at SLAC and NERSC, we compared its performance to that of bbcp and GridFTP and determined that they were comparable. With a proof of concept created using two four-node clusters with multiple distributed multi-core CPUs, network interfaces and flash memory, we achieved 155Gbps memory-to-memory over a 2x100Gbps link aggregated channel and 70Gbps file-to-file with encryption over a 5000 mile 100Gbps link.

97 MATHEMATICS AND COMPUTING↗

Efficient On-demand Connection Management Mechanisms with PGAS Models on InfiniBand

In the last decade or so, clusters have observed a tremendous rise in popularity due to the excellent price to performance ratio. A variety of Interconnects have been proposed during this period, with InfiniBand leading the way due to its high performance and open standard. At the same time, multiple programming models have emerged in order to meet the requirements of various applications and their programming models. To support requirements of multiple programming models, InfiniBand provides multiple transport semantics, ranging from unreliable connectionless to reliable connected characteristics. Among them, the reliable connection (RC) semantics is being widely due to its high performance and support for novel features like Remote Direct Memory Acesss (RDMA), hardware atomics and Network Fault Tolerance. However, the pairwise connection oriented nature of the RC transport semantics limits its scalability and usage at the increasing processor counts. In this paper, we design and implement on-demand connection management approaches in the context of Partitioned Global Address Space (PGAS) programming models, which provided shared memory abstraction and one-sided communication semantics, leading to the development of multiple languages (UPC, X10, Chapel) and libraries (Global Arrays, MPI-RMA). Using Global Arrays as the research vehicle, we implement this approach with Aggregate Remote Memory Copy Interface (ARMCI), the runtime system of Global Arrays. We evaluate our approach, ARMCI-On Demand Connection Management (ARMCI-ODCM) using various microbenchmarks and benchmarks (LU Factorization, RandomAccess and Lennard Jones simulation) and application (Subsurface transport over multiple phases (STOMP)). With the performance evaluation for up to 4096 processors, we are able to have a multi-fold reduction in connection memory with a negligible degradation in performance. Using STOMP at 4096 processors, reduces the overall connection memory by 66 times with no performance degradation. To the best of our knowledge, this is the first design, implementation and evaluation of on-demand connection management with InfiniBand using PGAS models.

54 ENVIRONMENTAL SCIENCES↗

A microbenchmark characterization of the Emu chick

The Emu Chick is a prototype system designed around the concept of migratory memory-side processing. Rather than transferring large amounts of data across power-hungry, high-latency interconnects, the Emu Chick moves lightweight thread contexts to near-memory cores before the beginning of each memory read. The current prototype hardware uses FPGAs to implement cache-less “Gossamer” cores for computational work and rely on a typical stationary core (PowerPC) to run basic operating system functions and migrate threads between nodes. In this multi-node characterization of the Emu Chick, we extend an earlier single-node investigation of the the memory bandwidth characteristics of the system through benchmarks like STREAM, pointer chasing, and sparse matrix-vector multiplication. Here, we compare the Emu Chick hardware to architectural simulation and an Intel Xeon-based platform. Our results demonstrate that for many basic operations the Emu Chick can use available memory bandwidth more efficiently than a more traditional, cache-based architecture although bandwidth usage suffers for computationally intensive workloads like SpMV. Moreover, the Emu Chick provides stable, predictable performance with up to 65% of the peak bandwidth utilization on a random-access pointer chasing benchmark with weak locality.

97 MATHEMATICS AND COMPUTING↗

Predictive model for real-time energy disaggregation using long short-term memory

To provide affordable energy-saving solutions for the small and medium-sized manufacturers (SMMs), we propose a unified framework for generating predictive models that support real-time disaggregation of power consumption from combined inputs, enabling automatic machine state identification simultaneously for joint analysis of energy usage patterns. Further, the proposed framework transforms raw power consumption into a time series with look-back and bootstrap capabilities for historical pattern detection, while a learning architecture utilizes the stacked long short-term memory (LSTM) layers as encoders for embedding generation with sequential awareness. Experimental results demonstrate 93.65% minimum accuracy in ideal case of real-time energy usage and machine state prediction.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

PeleC [SWR-17-35]

PeleC is an adaptive-mesh compressible hydrodynamics code for reacting flows. It solves the compressible Navier-Stokes with multispecies transport in a block structured framework. The resulting algorithm is well suited for flows with localized resolution requirements and robust to discontinuities. User controllable refinement crieteria has the potential to result in extremely small numerical dissipation and dispersion, making this code appropriate for both research and applied usage. The code is built on the AMReX library which facilitates hierarchical parallelism and manages distributed memory parallism. PeleC algorithms are implemented to express shared memory parallelism.

Source record↗

PELEC

PeleC is an adaptive-mesh compressible hydrodynamics code for reacting flows. It solves the compressible Navier-Stokes with multispecies transport in a block structured framework. The resulting algorithm is well suited for flows with localized resolution requirements and robust to discontinuities. User controllable refinement crieteria has the potential to result in extremely small numerical dissipation and dispersion, making this code appropriate for both research and applied usage. The code is built on the AMReX library which facilitates hierarchical parallelism and manages distributed memory parallism. PeleC algorithms are implemented to express shared memory parallelism.

Source record↗

Deploy Nalu/Kokkos algorithmic infrastructure with performance benchmarking.

The former Nalu interior heterogeneous algorithm design, which was originally designed to manage matrix assembly operations over all elemental topology types, has been modified to operate over homogeneous collections of mesh entities. This newly templated kernel design allows for removal of workset variable resize operations that were formerly required at each loop over a Sierra ToolKit (STK) bucket (nominally, 512 entities in size). Extensive usage of the Standard Template Library (STL) std::vector has been removed in favor of intrinsic Kokkos memory views. In this milestone effort, the transition to Kokkos as the underlying infrastructure to support performance and portability on many-core architectures has been deployed for key matrix algorithmic kernels. A unit-test driven design effort has developed a homogeneous entity algorithm that employs a team-based thread parallelism construct. The STK Single Instruction Multiple Data (SIMD) infrastructure is used to interleave data for improved vectorization. The collective algorithm design, which allows for concurrent threading and SIMD management, has been deployed for the core low-Mach element- based algorithm. Several tests to ascertain SIMD performance on Intel KNL and Haswell architectures have been carried out. The performance test matrix includes evaluation of both low- and higher-order methods. The higher-order low-Mach methodology builds on polynomial promotion of the core low-order control volume nite element method (CVFEM). Performance testing of the Kokkos-view/SIMD design indicates low-order matrix assembly kernel speed-up ranging between two and four times depending on mesh loading and node count. Better speedups are observed for higher-order meshes (currently only P=2 has been tested) especially on KNL. The increased workload per element on higher-order meshes bene ts from the wide SIMD width on KNL machines. Combining multiple threads with SIMD on KNL achieves a 4.6x speedup over the baseline, with assembly timings faster than that observed on Haswell architecture. The computational workload of higher-order meshes, therefore, seems ideally suited for the many-core architecture and justi es further exploration of higher-order on NGP platforms. A Trilinos/Tpetra-based multi-threaded GMRES preconditioned by symmetric Gauss Seidel (SGS) represents the core solver infrastructure for the low-Mach advection/diffusion implicit solves. The threaded solver stack has been tested on small problems on NREL's Peregrine system using the newly developed and deployed Kokkos-view/SIMD kernels. fforts are underway to deploy the Tpetra-based solver stack on NERSC Cori system to benchmark its performance at scale on KNL machines.

97 MATHEMATICS AND COMPUTING↗

Estimating Flexibility Envelopes for Residential Customers From Utility Smart Meter Data: Preprint

Demand response from residential customers has significant potential to support power system operations, but accurate flexibility estimation is challenging due to the limited resolution of advanced metering infrastructure (AMI) data. Most utility AMI measurements are recorded at hourly intervals, with only a small portion at higher resolutions, and even fewer households have appliance-level energy usage data. To address this issue, this paper proposes a two-stage long short-term memory (LSTM) framework for estimating household flexibility envelopes from low-resolution AMI data. In the first stage, the heating, ventilating, and air-conditioning (HVAC) load and non-HVAC loads are estimated by using a model trained on a small set of households with appliance-level profiles. These estimated data are then used to compute the upper- and lower-flexibility bounds, which are subsequently down-sampled to lower-resolution data. In the second stage, these flexibility bounds serve as training inputs for another LSTM model, enabling direct prediction of flexibility envelopes for households with only hourly AMI data. This method is validated using Pecan Street data from two different areas, and the results demonstrate its applicability and effectiveness.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Approximate but accurate quantum dynamics from the Mori formalism: I. Nonequilibrium dynamics

We present a formalism that explicitly unifies the commonly used Nakajima-Zwanzig approach for reduced density matrix dynamics with the more versatile Mori theory in the context of nonequilibrium dynamics. Employing a Dyson-type expansion to circumvent the difficulty of projected dynamics, we obtain a self-consistent equation for the memory kernel which requires only knowledge of normally evolved auxiliary kernels. To illustrate the properties of the current approach, we focus on the spin-boson model and limit our attention to the use of a simple and inexpensive quasi-classical dynamics, given by the Ehrenfest method, for the calculation of the auxiliary kernels. For the first time, we provide a detailed analysis of the dependence of the properties of the memory kernels obtained via different projection operators, namely, the thermal (Redfield-type) and population based (NIBA-type) projection operators. We further elucidate the conditions that lead to short-lived memory kernels and the regions of parameter space to which this program is best suited. Via a thorough analysis of the different closures available for the auxiliary kernels and the convergence properties of the self-consistently extracted memory kernel, we identify the mechanisms whereby the current approach leads to a significant improvement over the direct usage of standard semi- and quasi-classical dynamics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Lightweight storage and overlay networks for fault tolerance.

The next generation of capability-class, massively parallel processing (MPP) systems is expected to have hundreds of thousands to millions of processors, In such environments, it is critical to have fault-tolerance mechanisms, including checkpoint/restart, that scale with the size of applications and the percentage of the system on which the applications execute. For application-driven, periodic checkpoint operations, the state-of-the-art does not provide a scalable solution. For example, on today's massive-scale systems that execute applications which consume most of the memory of the employed compute nodes, checkpoint operations generate I/O that consumes nearly 80% of the total I/O usage. Motivated by this observation, this project aims to improve I/O performance for application-directed checkpoints through the use of lightweight storage architectures and overlay networks. Lightweight storage provide direct access to underlying storage devices. Overlay networks provide caching and processing capabilities in the compute-node fabric. The combination has potential to signifcantly reduce I/O overhead for large-scale applications. This report describes our combined efforts to model and understand overheads for application-directed checkpoints, as well as implementation and performance analysis of a checkpoint service that uses available compute nodes as a network cache for checkpoint operations.

99 GENERAL AND MISCELLANEOUS↗

A parallel, distributed memory implementation of the adaptive sampling configuration interaction method

The many-body simulation of quantum systems is an active field of research that involves several different methods targeting various computing platforms. Many methods commonly employed, particularly coupled cluster methods, have been adapted to leverage the latest advances in modern high-performance computing. Selected configuration interaction (sCI) methods have seen extensive usage and development in recent years. However, the development of sCI methods targeting massively parallel resources has been explored only in a few research works. Here, we present a parallel, distributed memory implementation of the adaptive sampling configuration interaction approach (ASCI) for sCI. In particular, we will address the key concerns pertaining to the parallelization of the determinant search and selection, Hamiltonian formation, and the variational eigenvalue calculation for the ASCI method. Load balancing in the search step is achieved through the application of memory-efficient determinant constraints originally developed for the ASCI-PT2 method. The presented benchmarks demonstrate near optimal speedup for ASCI calculations of Cr 2 (24e, 30o) with 10 6 , 10 7 , and 3 × 10 8 variational determinants on up to 16 384 CPUs. Importantly, to the best of the authors’ knowledge, this is the largest variational ASCI calculation to date.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Porting the WAVEWATCH III (v6.07) wave action source terms to GPU

Abstract. Surface gravity waves play a critical role in several processes, including mixing, coastal inundation, and surface fluxes. Despite the growing literature on the importance of ocean surface waves, wind–wave processes have traditionally been excluded from Earth system models (ESMs) due to the high computational costs of running spectral wave models. The development of the Next Generation Ocean Model for the DOE’s (Department of Energy) E3SM (Energy Exascale Earth System Model) Project partly focuses on the inclusion of a wave model, WAVEWATCH III (WW3), into E3SM. WW3, which was originally developed for operational wave forecasting, needs to be computationally less expensive before it can be integrated into ESMs. To accomplish this, we take advantage of heterogeneous architectures at DOE leadership computing facilities and the increasing computing power of general-purpose graphics processing units (GPUs). This paper identifies the wave action source terms, W3SRCEMD, as the most computationally intensive module in WW3 and then accelerates them via GPU. Our experiments on two computing platforms, Kodiak (P100 GPU and Intel(R) Xeon(R) central processing unit, CPU, E5-2695 v4) and Summit (V100 GPU and IBM POWER9 CPU) show respective average speedups of 2× and 4× when mapping one Message Passing Interface (MPI) per GPU. An average speedup of 1.4× was achieved using all 42 CPU cores and 6 GPUs on a Summit node (with 7 MPI ranks per GPU). However, the GPU speedup over the 42 CPU cores remains relatively unchanged (∼ 1.3×) even when using 4 MPI ranks per GPU (24 ranks in total) and 3 MPI ranks per GPU (18 ranks in total). This corresponds to a 35 %–40 % decrease in both simulation time and usage of resources. Due to too many local scalars and arrays in the W3SRCEMD subroutine and the huge WW3 memory requirement, GPU performance is currently limited by the data transfer bandwidth between the CPU and the GPU. Ideally, OpenACC routine directives could be used to further improve performance. However, W3SRCEMD would require significant code refactoring to make this possible. We also discuss how the trade-off between the occupancy, register, and latency affects the GPU performance of WW3.

58 GEOSCIENCES↗

Machine Learning-enabled Scalable Performance Prediction of Scientific Codes

Hardware architectures become increasingly complex as the compute capabilities grow to exascale. Here, we present the Analytical Memory Model with Pipelines (AMMP) of the Performance Prediction Toolkit (PPT). PPT-AMMP takes high-level source code and hardware architecture parameters as input and predicts runtime of that code on the target hardware platform, which is defined in the input parameters. PPT-AMMP transforms the code to an (architecture-independent) intermediate representation, then (i) analyzes the basic block structure of the code, (ii) processes architecture-independent virtual memory access patterns that it uses to build memory reuse distance distribution models for each basic block, and (iii) runs detailed basic-block level simulations to determine hardware pipeline usage. PPT-AMMP uses machine learning and regression techniques to build the prediction models based on small instances of the input code, then integrates into a higher-order discrete-event simulation model of PPT running on Simian PDES engine. We validate PPT-AMMP on four standard computational physics benchmarks and present a use case of hardware parameter sensitivity analysis to identify bottleneck hardware resources on different code inputs. We further extend PPT-AMMP to predict the performance of a scientific application code, namely, the radiation transport mini-app SNAP. To this end, we analyze multi-variate regression models that accurately predict the reuse profiles and the basic block counts. We validate predicted SNAP runtimes against actual measured times.

97 MATHEMATICS AND COMPUTING↗

tite

The TITE library provides type erasure implementation utilizing the tag_invoke paradigm proposed for standardization here: https://open-std.org/JTC1/SC22/WG21/docs/papers/2019/p1895r0.pdf. The implementation contained herein is largely modeled after that provided in the standardization proposal and available at https://godbolt.org/z/3TvO4f. Significant modification have been made to the original implementation to improve its suitability to be utilized for GPU architectures. In particular the implementation: - only requires a C++14 standard. - has been extended to obtain vtables for GPU device architectures in addition to CPU host architectures. - provides a gpu_allocator class appropriate for allocation of the type-erased object to GPU memory - provides copy semantics omitted from the original implementation-- - generally the copy semantics are to completely copy the type-erased object - when the usage of the type-erased object guarantees immutability the copy semantics are altered to reference-counted shallow copies (copies of pointers) for improved performance

Solomon, CJ↗

L-BFGS Class Implementation in C++

This report presents a header-only C++ class implementation of the Limited-memory BroydenFletcher-Goldfarb-Shanno (L-BFGS) algorithm. The L-BFGS method is a general purpose quasi-Netwon optimization method that builds an approximation of the descent direction from consecutive iterate and gradient vectors. The limited-memory aspect stems from the replacement of the N × N approximation matrix of the original BFGS method with M vectors of length N. An example usage of the class is included along with the reference source code.

97 MATHEMATICS AND COMPUTING↗