Search NASA⌕ Search

SEARCH · Search NASA

Results for “memory efficiency”

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 451 records · Page 25

An integrated runtime and compile-time approach for parallelizing structured and block structured applications

Scientific and engineering applications often involve structured meshes. These meshes may be nested (for multigrid codes) and/or irregularly coupled (called multiblock or irregularly coupled regular mesh problems). A combined runtime and compile-time approach for parallelizing these applications on distributed memory parallel machines in an efficient and machine-independent fashion was described. A runtime library which can be used to port these applications on distributed memory machines was designed and implemented. The library is currently implemented on several different systems. To further ease the task of application programmers, methods were developed for integrating this runtime library with compilers for HPK-like parallel programming languages. How this runtime library was integrated with the Fortran 90D compiler being developed at Syracuse University is discussed. Experimental results to demonstrate the efficacy of our approach are presented. A multiblock Navier-Stokes solver template and a multigrid code were experimented with. Our experimental results show that our primitives have low runtime communication overheads. Further, the compiler parallelized codes perform within 20 percent of the code parallelized by manually inserting calls to the runtime library.

Agrawal, Gagan↗

Simplified Parallel Domain Traversal

Many data-intensive scientific analysis techniques require global domain traversal, which over the years has been a bottleneck for efficient parallelization across distributed-memory architectures. Inspired by MapReduce and other simplified parallel programming approaches, we have designed DStep, a flexible system that greatly simplifies efficient parallelization of domain traversal techniques at scale. In order to deliver both simplicity to users as well as scalability on HPC platforms, we introduce a novel two-tiered communication architecture for managing and exploiting asynchronous communication loads. We also integrate our design with advanced parallel I/O techniques that operate directly on native simulation output. We demonstrate DStep by performing teleconnection analysis across ensemble runs of terascale atmospheric CO{sub 2} and climate data, and we show scalability results on up to 65,536 IBM BlueGene/P cores.

99 GENERAL AND MISCELLANEOUS↗

Nonvolatile Memories in Spiking Neural Network Architectures: Current and Emerging Trends

A sustainable computing scenario demands more energy-efficient processors. Neuromorphic systems mimic biological functions by employing spiking neural networks for achieving brain-like efficiency, speed, adaptability, and intelligence. Current trends in neuromorphic technologies address the challenges of investigating novel materials, systems, and architectures for enabling high-integration and extreme low-power brain-inspired computing. This review collects the most recent trends in exploiting the physical properties of nonvolatile memory technologies for implementing efficient in-memory and in-device computing with spike-based neuromorphic architectures.

42 ENGINEERING↗

Inflated speedups in parallel simulations via malloc()

Discrete-event simulation programs make heavy use of dynamic memory allocation in order to support simulation's very dynamic space requirements. When programming in C one is likely to use the malloc() routine. However, a parallel simulation which uses the standard Unix System V malloc() implementation may achieve an overly optimistic speedup, possibly superlinear. An alternate implementation provided on some (but not all systems) can avoid the speedup anomaly, but at the price of significantly reduced available free space. This is especially severe on most parallel architectures, which tend not to support virtual memory. It is shown how a simply implemented user-constructed interface to malloc() can both avoid artificially inflated speedups, and make efficient use of the dynamic memory space. The interface simply catches blocks on the basis of their size. The problem is demonstrated empirically, and the effectiveness of the solution is shown both empirically and analytically.

Nicol, David M.↗

MIND-MAC: Multi-Level In-memory Quasi Non-Destructive MAC Operation in Compact 2T-nC FeRAM for Efficient DNN Accelerator

We present MIND-MAC, a compact 2T-nC FeRAM architecture that performs multi-level, quasi-non-destructive in-memory multiply–accumulate (MAC) for deep neural networks. By exploiting voltage-controlled partial domain switching in MFM capacitors and read-transistor amplification, the cell stores multi-bit weights and gates bit-serial inputs to produce an accumulated current on shared lines. We combine TCAD-extracted parasitics with experimentally calibrated ferroelectric models in SPICE to validate device-/circuit-level behavior, and validate multi-level sensing and QNRO with measurements on a fabricated 2T-3C test vehicle. An analytical system model maps MIND-MAC to a 6-GB main-memory in-memory compute (IMC) architecture and benchmarks VGG13 inference in 61.08 ms at 964.99 mJ. Results indicate high density, reduced rewrite overhead, and energy efficiency, positioning 2T-nC FeRAM as a promising IMC candidate for next-generation AI hardware.

36 MATERIALS SCIENCE↗

Multi-Dimensional Spectral Difference Method for Unstructured Grids

A new, high-order, conservative, and efficient method for conservation laws on unstructured grids is developed. It combines the best features of structured and unstructured grid methods to attain computational efficiency and geometric flexibility; it utilizes the concept of discontinuous and high-order local representations to achieve conservation and high accuracy; and it is based on the finite-difference formulation for simplicity. Universal reconstructions are obtained by distributing unknowns in a geometrically similar manner for all unstructured cells. Placements of the unknown and flux points with various order of accuracy are given for the line, triangular and tetrahedral elements. The data structure of the new method permits an optimum use of cache memory, resulting in further computational efficiency on modern computers. A new pointer system is developed that reduces memory requirements and simplifies programming for any order of accuracy. Numerical solutions are presented and compared with the exact solutions for wave propagation problems in both two and three dimensions to demonstrate the capability of the method. Excellent agreement has been found. The method is simpler and more efficient than previous discontinuous Galerkin and spectral volume methods for unstructured grids.

Liu, Yen↗

Time-Sequenced Flow Field Prediction in an Optima Spark-Ignition Direct-Injection Engine Using Bidirectional Recurrent Neural Network (bi-RNN) with Long Short-Term Memory

To further improve the energy conversion efficiency of internal combustion engine, the transient and complex air flow movement inside the cylinder needs to be better understood and controlled. Although the in-cylinder flow fields are highly stochastic with strong cycle-to-cycle fluctuations, machine learning can still provide an efficient way to learn and regress the complex flow movement process inside the cylinder. In this work, a bidirectional recurrent neural network (bi-RNN) model with long short-term memory was applied to predict the in-cylinder flow fields at different time steps using training data from mull-cycle particle image velocimetry (PIV) measurements. To evaluate the agreement between the true and predicted flow fields, structure and magnitude comparison indices are calculated both globally and locally. The comparison results show that the bi-RNN model can accurately predict the bulk flow and vortex motions from early intake stroke to compression stroke. This work demonstrates that the machine learning model has the potential to predict the underlying dynamics of the interaction between in-cylinder flows and provides a reliable way to improve temporal resolution in PIV flow data to better reveal transient in-cylinder flow features.

Bi-RNN model↗

Speeding up and reducing memory usage for scientific machine learning via mixed precision

Scientific machine learning (SciML) has emerged as a versatile approach to address complex computational science and engineering problems. Within this field, physics-informed neural networks (PINNs) and deep operator networks (DeepONets) stand out as the leading techniques for solving partial differential equations by incorporating both physical equations and experimental data. However, training PINNs and DeepONets require significant computational resources, including long computational times and large amounts of memory. In search of computational efficiency, training neural networks using half precision (float16) rather than the conventional single (float32) or double (float64) precision has gained substantial interest, given the inherent benefits of reduced computational time and memory consumed. However, we find that float16 cannot be applied to SciML methods, because of gradient divergence at the start of training, weight updates going to zero, and the inability to converge to a local minima. To overcome these limitations, we explore mixed precision, which is an approach that combines the float16 and float32 numerical formats to reduce memory usage and increase computational speed. Our experiments showcase that mixed precision training not only substantially decreases training times and memory demands but also maintains model accuracy. Here, we also reinforce our empirical observations with a theoretical analysis. The research has broad implications for SciML in various computational applications.

97 MATHEMATICS AND COMPUTING↗

Integration of Ag-CBRAM crossbars and Mott ReLU neurons for efficient implementation of deep neural networks in hardware

In-memory computing with emerging non-volatile memory devices (eNVMs) has shown promising results in accelerating matrix-vector multiplications. However, activation function calculations are still being implemented with general processors or large and complex neuron peripheral circuits. Here, we present the integration of Ag-based conductive bridge random access memory (Ag-CBRAM) crossbar arrays with Mott rectified linear unit (ReLU) activation neurons for scalable, energy and area-efficient hardware (HW) implementation of deep neural networks. We develop Ag-CBRAM devices that can achieve a high ON/OFF ratio and multi-level programmability. Compact and energy-efficient Mott ReLU neuron devices implementing ReLU activation function are directly connected to the columns of Ag-CBRAM crossbars to compute the output from the weighted sum current. We implement convolution filters and activations for VGG-16 using our integrated HW and demonstrate the successful generation of feature maps for CIFAR-10 images in HW. Our approach paves a new way toward building a highly compact and energy-efficient eNVMs-based in-memory computing system.

Mott insulators↗

Analysis and mitigation of parasitic resistance effects for analog in-memory neural network acceleration

To support the increasing demands for efficient deep neural network processing, accelerators based on analog in-memory computation of matrix multiplication have recently gained significant attention for reducing the energy of neural network inference. However, analog processing within memory arrays must contend with the issue of parasitic voltage drops across the metal interconnects, which distort the results of the computation and limit the array size. This work analyzes how parasitic resistance affects the end-to-end inference accuracy of state-of-the-art convolutional neural networks, and comprehensively studies how various design decisions at the device, circuit, architecture, and algorithm levels affect the system's sensitivity to parasitic resistance effects. Here, a set of guidelines are provided for how to design analog accelerator hardware that is intrinsically robust to parasitic resistance, without any explicit compensation or re-training of the network parameters.

97 MATHEMATICS AND COMPUTING↗

Scalable PGAS Metadata Management on Extreme Scale Systems

Programming models intended to run on exascale systems have a number of challenges to overcome, specially the sheer size of the system as measured by the number of concurrent software entities created and managed by the underlying runtime. It is clear from the size of these systems that any state maintained by the programming model has to be strictly sub-linear in size, in order not to overwhelm memory usage with pure overhead. A principal feature of Partitioned Global Address Space (PGAS) models is providing easy access to global-view distributed data structures. In order to provide efficient access to these distributed data structures, PGAS models must keep track of metadata such as where array sections are located with respect to processes/threads running on the HPC system. As PGAS models and applications become ubiquitous on very large transpetascale systems, a key component to their performance and scalability will be efficient and judicious use of memory for model overhead (metadata) compared to application data. We present an evaluation of several strategies to manage PGAS metadata that exhibit different space/time tradeoffs. We use two real-world PGAS applications to capture metadata usage patterns and gain insight into their communication behavior.

Global Arrays↗

Quantum optical memory for entanglement distribution

Optical photons are powerful carriers of quantum information, which can be delivered in free space by satellites or in fibers on the ground over long distances. Entanglement of quantum states over long distances can empower quantum computing, quantum communications, and quantum sensing. Quantum optical memories are devices designed to store quantum information in the form of stationary excitations, such as atomic coherence, and are capable of coherently mapping these excitations to flying qubits. Quantum memories can effectively store and manipulate quantum states, making them indispensable elements in future long-distance quantum networks. Over the past two decades, quantum optical memories with high fidelities, high efficiencies, long storage times, and promising multiplexing capabilities have been developed, especially at the single-photon level. In this review, we introduce the working principles of commonly used quantum memory protocols and summarize the recent advances in quantum memory demonstrations. We also offer a vision for future quantum optical memory devices that may enable entanglement distribution over long distances.

Lei, Yisheng↗

Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training

Deep Learning Recommendation Models (DLRMs) are critical applications in various domains and have evolved as one of the single largest machine learning applications. Trillions of DLRM parameters exceed the on-chip memory capacity of GPUs. Large-scale multi-node systems are required for distributed DLRM inference and training, which suffer from the all-to-all communication bottleneck, mainly limiting the scalability of ever-growing DLRMs. In recent years, SmartNICs have evolved with coupled computation and communication capabilities providing opportunities for a powerful heterogeneous device in the system. However, there isn't such a distributed system that fully leverages the abundant smartNIC resources that resolve the scalability issue of DLRMs. In this work, we proposed a software-hardware co-design of a heterogeneous smartNIC system that resolves the communication bottleneck of distributed DLRMs, mitigates the memory bandwidth pressure, and improves computation efficiency. We provide a set of smartNIC designs of cache systems (including local cache and remote cache) and smartNIC computation kernels which reduce data movement, relieve memory lookup intensity, and improve the GPU's computation efficiency. In addition, we propose a graph algorithm that improves the data locality of queries within batches which optimizes the overall system performance with higher data reuse. Our evaluation shows that our system achieves 2.1x latency speedup for inference and 1.6x throughput speedup for training.

Guo, Anqi↗

SBLLmalloc V1.0

Memory size has long limited large-scale applications on high-performance computing (HPC) systems. Increasing core counts per chip and power density constraints, which limit the number of DIMMs per node, have exacerbated this problem. Mechanisms to manage memory usage - preferably transparently - more efficiently could increase effective DRAM capacity and, thus, the benefit of multicore similarity. These data regions occupy multiple physical locations within a multicore node and thus offer a potential savings in memory capacity. These data, primarily residing in heap, are quite dynamic, which makes them difficult to manage statically. SBLLmalloc is a memory allocation library that automatically identifies the replicated memory blocks and merges them into a single copy. SBLLmalloc does not require application or OS changes since we implement it as a user-level library that can be linked at runtime. SBLLmalloc provides the capability to run larger problems using the same hardware configuration in multicore nodes. SBLLmalloc has been implemented for Linux clusters.

Source record↗

SBLLmalloc V1.0

Memory size has long limited large-scale applications on high-performance computing (HPC) systems. Increasing core counts per chip and power density constraints, which limit the number of DIMMs per node, have exacerbated this problem. Mechanisms to manage memory usage - preferably transparently - more efficiently could increase effective DRAM capacity and, thus, the benefit of multicore similarity. These data regions occupy multiple physical locations within a multicore node and thus offer a potential savings in memory capacity. These data, primarily residing in heap, are quite dynamic, which makes them difficult to manage statically. SBLLmalloc is a memory allocation library that automatically identifies the replicated memory blocks and merges them into a single copy. SBLLmalloc does not require application or OS changes since we implement it as a user-level library that can be linked at runtime. SBLLmalloc provides the capability to run larger problems using the same hardware configuration in multicore nodes. SBLLmalloc has been implemented for Linux clusters.

Source record↗

Evaluating HPC Kernels for Processing in Memory

Memory subsystems contribute significantly to the performance and energy efficiency of high-performance computing (HPC) applications. Traditional memory technologies with conventional organization (e.g., DRAM) are struggling to keep up with the increasing memory requirements of modern applications. Techniques such as multilayer cache hierarchy and out-of-order execution are still falling short of mitigating the penalty incurred by memory accesses. Processing-in-memory (PIM), which involves moving memory-intensive kernels to memory for execution instead of bringing the data to the processing unit, is emerging as a promising technique. PIM has recently received traction among computer architecture researchers, and the increasing research activity surrounding this technique indicates its potential to alleviate main memory performance bottlenecks. In this paper, we characterize and identify memory-intensive HPC kernels, perform a first-order evaluation of the PIM technique for selected HPC kernels, quantify performance deviation, and analyze the key factors that affect PIM efficiency.

Asifuzzaman, Kazi↗

FISH: A THREE-DIMENSIONAL PARALLEL MAGNETOHYDRODYNAMICS CODE FOR ASTROPHYSICAL APPLICATIONS

FISH is a fast and simple ideal magnetohydrodynamics code that scales to {approx}10,000 processes for a Cartesian computational domain of {approx}1000{sup 3} cells. The simplicity of FISH has been achieved by the rigorous application of the operator splitting technique, while second-order accuracy is maintained by the symmetric ordering of the operators. Between directional sweeps, the three-dimensional data are rotated in memory so that the sweep is always performed in a cache-efficient way along the direction of contiguous memory. Hence, the code only requires a one-dimensional description of the conservation equations to be solved. This approach also enables an elegant novel parallelization of the code that is based on persistent communications with MPI for cubic domain decomposition on machines with distributed memory. This scheme is then combined with an additional OpenMP parallelization of different sweeps that can take advantage of clusters of shared memory. We document the detailed implementation of a second-order total variation diminishing advection scheme based on flux reconstruction. The magnetic fields are evolved by a constrained transport scheme. We show that the subtraction of a simple estimate of the hydrostatic gradient from the total gradients can significantly reduce the dissipation of the advection scheme in simulations of gravitationally bound hydrostatic objects. Through its simplicity and efficiency, FISH is as well suited for hydrodynamics classes as for large-scale astrophysical simulations on high-performance computer clusters. In preparation for the release of a public version, we demonstrate the performance of FISH in a suite of astrophysically orientated test cases.

79 ASTRONOMY AND ASTROPHYSICS↗

The Reverse Cuthill-McKee Algorithm in Distributed-Memory

Ordering vertices of a graph is key to minimize fill-in and data structure size in sparse direct solvers, maximize locality in iterative solvers, and improve performance in graph algorithms. Except for naturally parallelizable ordering methods such as nested dissection, many important ordering methods have not been efficiently mapped to distributed-memory architectures. In this paper, we present the first-ever distributed-memory implementation of the reverse Cuthill-McKee (RCM) algorithm for reducing the profile of a sparse matrix. Our parallelization uses a two-dimensional sparse matrix decomposition. We achieve high performance by decomposing the problem into a small number of primitives and utilizing optimized implementations of these primitives. Our implementation shows strong scaling up to 1024 cores for smaller matrices and up to 4096 cores for larger matrices.

97 MATHEMATICS AND COMPUTING↗