An Energy-Efficient Reconfigurable Mobile Memory Interface for Computing Systems

Three-dimensional Light Detection and Ranging (3D LiDAR) sensors are widely used in autonomous vehicles including localization, sensing, and mapping. However, their significant computational requirements remains a major limitation. Although high-end Graphics Processing Units (GPUs) and Central Processing Units (CPUs) can solve the aforementioned problem, the cost and power consumption hinder the widespread adoption of 3-D LiDAR in commercial vehicles. To overcome these limitations, this paper presents a novel high-efficiency 3D mapping method that aims to achieve faster computation speeds than CPUs while providing better energy efficiency compared to GPUs. First, an optimized real-time LiDAR data preprocessing procedure is proposed, which is based on a fixed-point computation framework. Second, to accelerate the LiDAR point cloud matching process, we introduce a new processing architecture called Hardware Accelerated Brute Force Nearest Neighbor (HA-BFNN). Third, we further enhance real-time performance and energy efficiency by proposing a streaming FPGA accelerator architecture for both HA-BFNN and the Iterative Closest Point (ICP) algorithm. Experimental results show that our custom test board, based on the AMD Kintex-7 chip, accelerates point cloud matching, completing single-frame calculations in 5.76 ms, significantly outperforming other 3D mapping implementations on both CPUs and GPUs. Moreover, our FPGA implementation achieves up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$17.36\times $</tex-math> </inline-formula> speedup in execution time compared to CPU implementations. Finally, the proposed system enables real-time performance while consuming only 3.4W of power, maintaining accuracy comparable to software counterparts and even state-of-the-art 3D mapping methods.

For the past 40 years, Moore’s law has predicted the rapid growth of the computer industry. In the past few years, however, this growth has slowed for central processing units (CPUs). Instead, there has been a shift to multicore computing, specifically with the general purpose graphic processing units (GPUs). Conventional CPUs have between two and eight cores, but the GPUs can have hundreds, even thousands of cores. By parallelizing code, the computing power of these cores can be utilized to greatly accelerate the performance of certain algorithms. The GPUs, however, have been known to consume more power than the conventional processing units. While previous research has focused on the impact that the GPUs have on performance, there are much fewer studies on the impact of GPUs on energy consumption and efficiency. Some researchers have hypothesized that if the performance of an algorithm was sufficiently increased on a GPU, then the accelerated time would actually cause the GPU to consume less energy. For the first time to our knowledge, we study the energy efficiency of a GPU with an application to iris recognition. Using GPU-based code written in the C++ compatible compute unified device architecture language, energy consumption tests are performed on basic image processing techniques, including image inversion, thresholding, dilation, erosion, and memory/computationally intensive calculations, such as the template matching. We demonstrate that the portions of these algorithms implemented on the GPU reduce energy consumption by as much as 272 times.

SummaryMany e‐business or social network servers have been constructed on cloud. On such open environments, private data of massive users have to be protected by encrypting, such as using Advanced Encryption Standard (AES), and furthermore, this process must be finished in a short time for users' better experience. This gives huge pressure on cloud servers, especially common servers, such as web servers. We urgently need an inexpensive and highly efficient method to relieve cloud servers' pressure. Fortunately, many cores of a graphics processing unit (GPU) can undertake this hard mission because of stronger computing power and lower price. The GPU environments can be virtualized on demand by cloud through the vCUDA technology. Of course, for those clouds not equipped with a GPU, a central processing unit (CPU) can still work as multithreads in parallel. Thus, in a cloud, AES can be parallelized using many cores of a GPU or multicores of a CPU with high efficiency and low cost. For typical cloud applications, such as web services, there are massive users and each one has short plaintext. If we simply parallelize AES in such an application, we cannot obtain better performance because of the GPU's extra data transferring cost. Thus, we coalesce the massive users' data and cut these data into same‐length slices for improving the performance of parallel AES as much as possible. So we design six parallel AES algorithms using GPU parallelism or CPU parallelism, which differ in parallel scope and whether data are coalesced or cut to slices. Specifically, they are coalescent and sliced GPU (GCS), coalescent and unsliced GPU, uncoalescent GPU, coalescent and sliced CPU, coalescent and unsliced CPU, and uncoalescent CPU. Moreover, we implement them on two representative platforms and evaluate their performance. Through comparing their performance, GCS has the best performance among these algorithms. In a cloud with Nvidia GPUs, GCS is a more powerful algorithm for massive users' data encrypting, relatively. Copyright © 2015 John Wiley &amp; Sons, Ltd.

Efficient methods for implementation of multi-level nonrigid mass-preserving image registration on GPUs and multi-threaded CPUs

We have implemented the Martini force field within Lawrence Livermore National Laboratory's molecular dynamics program, ddcMD. The program is extended to a heterogeneous programming model so that it can exploit graphics processing unit (GPU) accelerators. In addition to the Martini force field being ported to the GPU, the entire integration step, including thermostat, barostat, and constraint solver, is ported as well, which speeds up the simulations to 278-fold using one GPU vs one central processing unit (CPU) core. A benchmark study is performed with several test cases, comparing ddcMD and GROMACS Martini simulations. The average performance of ddcMD for a protein-lipid simulation system of 136k particles achieves 1.04 µs/day on one NVIDIA V100 GPU and aggregates 6.19 µs/day on one Summit node with six GPUs. The GPU implementation in ddcMD offloads all computations to the GPU and only requires one CPU core per simulation to manage the inputs and outputs, freeing up remaining CPU resources on the compute node for alternative tasks often required in complex simulation campaigns. The ddcMD code has been made open source and is available on GitHub at https://github.com/LLNL/ddcMD.

MSA Probs is a parallel algorithm developed to align multiple sequence alignment using a central processing unit (CPU). Whereas the CPU has some limitations, such as the inability to parallelize tasks in the processor (latency-oriented). To overcome these limitations, this paper proposes an improved version of MSA Probs that is compatible with a graphical Processing Unit (GPU). This idea helps in enhancing the performance of our algorithm (MSAprobs). To parallelize the sequential algorithm, Compute Unified Device Architecture (CUDA) or OpenCL is commonly used on GPUs. The NIVIDIA API is used to investigate the GPU's computing power. The results of using CPU only in MSAprobs versus the CPU and GPU are compared using two data sets from the Bali Base and OX Bench. The evaluation of the CPU and GPU is done using Threads 1,2 and 4. The results showed that by combining CPU and GPU, performance is improved and execution time is reduced.

This study aims to employ both central processing units (CPUs) and graphics processing units (GPUs) to collaboratively generate contour lines from a large-scale digital elevation model (DEM). The performance was improved with regard to three aspects. First, the original DEM data were decomposed and assigned according to the GPU’s limited memory so that large-scale data could be correctly addressed. Second, different types of computational tasks between the CPUs and GPUs were dynamically scheduled to ensure that both accelerators cooperate for performance improvement. Third, parallel processing on GPUs and CPUs was separately optimized for more efficient acceleration. Experimental results indicated that applying the parallel algorithm to data with a volume of 37.81 GB and area of 5,975,625.16 km2 reduced the total execution time from 332.84 min to 8.29 min for an optimal speedup of 40.15. In addition, we investigated the effects of the computational intensity, decomposition granularity, and task scheduling on parallel efficiency and performance improvement. We also discussed its degree of effectiveness, broader application, and the future direction of research in the field of geographic information systems.

GPU acceleration of MPAS microphysics WSM6 using OpenACC directives: Performance and verification

As computation schemes evolve and many new tools become available to programmers to enhance the performance of their applications, many programmers started to look towards highly parallel platforms such as Graphical Processing Unit (GPU). Offloading computations that can take advantage of the architecture of the GPU is a technique that has proven fruitful in recent years. This technology enhances the speed and responsiveness of applications. Also, as a side effect, it reduces the power requirements for those applications and therefore extends portable devices battery life and helps computing clusters to run more power efficiently. Many performance analysis tools such as LTTng, strace and SystemTap already allow Central Processing Unit (CPU) tracing and help programmers to use CPU resources more efficiently. On the GPU side, different tools such as Nvidia’s Nsight, AMD’s CodeXL, and third party TAU and VampirTrace allow tracing Application Programming Interface (API) calls and OpenCL kernel execution. These tools are useful but are completely separate, and none of them allow a unified CPU-GPU tracing experience. We propose an extension to the existing scalable and highly efficient LTTng tracing platform to allow unified tracing of GPU along with CPU’s full tracing capabilities.

Purpose:The calculation time is a trade‐off for improving the accuracy of convolution dose calculation with fine calculation spacing of the KERMA kernel. We investigated to accelerate the convolution calculation using an ideal kernel on the Graphic Processing Units (GPU).Methods:The calculation was performed on the AMD graphics hardware of Dual FirePro D700 and our algorithm was implemented using the Aparapi that convert Java bytecode to OpenCL. The process of dose calculation was separated with the TERMA and KERMA steps. The dose deposited at the coordinate (x, y, z) was determined in the process. In the dose calculation running on the central processing unit (CPU) of Intel Xeon E5, the calculation loops were performed for all calculation points. On the GPU computation, all of the calculation processes for the points were sent to the GPU and the multi‐thread computation was done. In this study, the dose calculation was performed in a water equivalent homogeneous phantom with 1503 voxels (2 mm calculation grid) and the calculation speed on the GPU to that on the CPU and the accuracy of PDD were compared.Results:The calculation time for the GPU and the CPU were 3.3 sec and 4.4 hour, respectively. The calculation speed for the GPU was 4800 times faster than that for the CPU. The PDD curve for the GPU was perfectly matched to that for the CPU.Conclusion:The convolution calculation with the ideal kernel on the GPU was clinically acceptable for time and may be more accurate in an inhomogeneous region. Intensity modulated arc therapy needs dose calculations for different gantry angles at many control points. Thus, it would be more practical that the kernel uses a coarse spacing technique if the calculation is faster while keeping the similar accuracy to a current treatment planning system.

For memory design, it will go through a series of quality assurance checks to validate the completeness of the collaterals before releasing to production usage. The quality check for memory collateral is similar to the quality check done within standard cell collaterals. The issue with memory collateral validation is the IP coverage, as memory instantiation is manual, unlike standard cell instantiation is handled by the EDA tool automatically. Hence, for memory design, the permutation of memory cells instantiation, the permutation of placement within the same type of memory IP and across memory IP types as well as exhaustive coverage of memory collateral from timing, layout, noise, functional model and reliability model are always the question unanswered, prior production release. Prior production release of memory collateral, limited memory instantiation is done based on design instantiation, such as application processing unit (APU) design blocks, digital signal processing (DSP) design block, graphic processing unit (GPU) design blocks as well as central processing unit (CPU) design blocks. For other hardmacro IPs such as GPIO or SERDES, the IP coverage is not a concern as the IP variant is limited. This paper will share the general-purpose memory design for the memory compiler’s pre-silicon validation that addresses the above concerns. Hopefully, the sharing will benefit the design community.

Implementation of Image Enhancement Algorithms and Recursive Ray Tracing using CUDA

Watershed analysis, as a fundamental component of digital terrain analysis, is based on the Digital Elevation Model (DEM), which is a grid (raster) model of the Earth surface and topography. Watershed analysis consists of computationally and data intensive computing algorithms that need to be implemented by leveraging parallel and high-performance computing methods and techniques. In this paper, the Multiple Flow Direction (MFD) algorithm for watershed analysis is implemented and evaluated on multi-core Central Processing Units (CPU) and many-core Graphics Processing Units (GPU), which provides significant improvements in performance and energy usage. The implementation is based on NVIDIA CUDA (Compute Unified Device Architecture) implementation for GPU, as well as on OpenACC (Open ACCelerators), a parallel programming model, and a standard for parallel computing. Both phases of the MFD algorithm (i) iterative DEM preprocessing and (ii) iterative MFD algorithm, are parallelized and run over multi-core CPU and GPU. The evaluation of the proposed solutions is performed with respect to the execution time, energy consumption, and programming effort for algorithm parallelization for different sizes of input data. An experimental evaluation has shown not only the advantage of using OpenACC programming over CUDA programming in implementing the watershed analysis on a GPU in terms of performance, energy consumption, and programming effort, but also significant benefits in implementing it on the multi-core CPU.

Abstract. The advent of heterogeneous supercomputers with multi-core central processing units (CPUs) and graphics processing units (GPUs) requires geoscientific codes to be adapted to these new architectures. Here, we describe the porting of Meso-NH version 5.5 community weather research code to GPUs named MESONH-v55-OpenACC with guaranteed bit reproducibility thanks to its own MPPDB_CHECK library. This porting includes the use of OpenACC directives, specific memory management, communications optimization, development of a geometric multigrid (MG) solver and creation of an in-house preprocessor. Performance on the AMD MI250X GPU Adastra platform shows up to 6.0× speedup (4.6× on the NVIDIA A100 Leonardo platform), and achieves a gain of a factor of 2.3 in energy efficiency compared to AMD Genoa CPU Adastra platform using the same configuration with 64 nodes. The code is even 17.8× faster by halving the precision and quadrupling the nodes with a gain in energy efficiency of a factor of 1.3. First scientific simulations of three representative storms using 128 GPUs nodes of Adastra show successful cascade of scales for horizontal grid spacing down to 100 m and grid size up to 2.1 billion points. For one of these storms, Meso-NH is also successfully coupled to the WAVEWATCH III wave model via the OASIS3-MCT coupler without any extra computational cost. This GPU porting paves the way for Meso-NH to be used on future European exascale machines.

We present a novel heterogeneous parallel matrix multiplication algorithm that utilizes both central processing units (CPUs) and graphics processing units (GPUs) for large-scale matrices. Based on Strassen’s method, we represent matrix multiplication work as a set of matrix addition and multiplication tasks among their sub-matrices. Then, we distribute the tasks to CPUs and GPUs while considering the characteristics of the tasks and computing resources to minimize the data communication overhead and fully utilize the available computing power. To handle a large matrix efficiently with limited GPU memory, we also propose a block-based work decomposition method. We then further improve the performance of our method by exploiting the concurrent execution abilities of a heterogeneous parallel computing system. We implemented our method on five different heterogeneous systems and applied it to matrices of various sizes. Our method generally shows higher performance than the prior GPU-based matrix multiplication methods. Moreover, compared with the state-of-the-art GPU matrix multiplication library (i.e., CUBLAS), our method achieved up to 1.97 times higher performance using the same GPUs and CPU cores. In some cases, our method using a low-performance GPU (e.g., GTX 1060, 3 GB) achieved performance comparable to that of CUBLAS using a high-performance GPU (e.g., RTX 2080, 8 GB). Also, our method continually improves performance as we use more computing resources like additional CPU cores and GPUs. We could achieve such high performance because our approach fully utilized the capacities of the given heterogeneous parallel computing systems while employing the Strassen’s method, which has a lower asymptotic complexity. These results demonstrate the efficiency and robustness of our algorithm.