Towards Efficient Short-Range Pair Interaction on Sunway Many-Core Architecture

The short-range pair interaction consumes most of the CPU time in molecular dynamics(MD)simulations.The inherent computation sparsity makes it challenging to achieve high-performance kernel on the emerging many-core ar-chitecture.In this paper,we present a highly efficient short-range force kernel on the Sunway,a novel many-core architecture with many unique features.The parallel efficiency of this algorithm on the Sunway many-core processor is strongly limited by the poor data locality and write conflicts.To enhance the data locality,we adopt a super cluster based neighbor list with an appropriate granularity that fits in the local memory of computing cores.In the absence of a low overhead locking mechanism,using data-privatization force array is a more feasible method to avoid write conflicts,but results in the large overhead of data reduction.We adopt a dual-slice partitioning scheme for both hardware resources and computing tasks,which utilizes the on-chip data communication to reduce data reduction overhead and provide load balancing.Moreover,we exploit the single instruction multiple data(SIMD)parallelism and perform instruction reordering of the force kernel on this many-core processor.The experimental results show that the optimized force kernel obtains a performance speedup of 226x compared with the reference implementation and achieves 20％of peak flop rate on the Sunway many-core processor.     

Unstructured computational aerodynamics on many integrated core architecture

Shared memory parallelization of the flux kernel of PETSc-FUN3D, an unstructured tetrahedral mesh Euler flow code previously studied for distributed memory and multi-core shared memory, is evaluated on up to 61 cores per node and up to 4 threads per core. We explore several thread-level optimizations to improve flux kernel performance on the state-of-the-art many integrated core (MIC) Intel processor Xeon Phi “Knights Corner,” with a focus on strong thread scaling. While the linear algebraic kernel is bottlenecked by memory bandwidth for even modest numbers of cores sharing a common memory, the flux kernel, which arises in the control volume discretization of the conservation law residuals and in the formation of the preconditioner for the Jacobian by finite-differencing the conservation law residuals, is compute-intensive and is known to exploit effectively contemporary multi-core hardware. We extend study of the performance of the flux kernel to the Xeon Phi in three thread affinity modes, namely scatter, compact, and balanced, in both offload and native mode, with and without various code optimizations to improve alignment and reduce cache coherency penalties. Relative to baseline “out-of-the-box” optimized compilation, code restructuring optimizations provide about 3.8x speedup using the offload mode and about 5x speedup using the native mode. Even with these gains for the flux kernel, with respect to execution time the MIC simply achieves par with optimized compilation on a contemporary multi-core Intel CPU, the 16-core Sandy Bridge E5 2670. Nevertheless, the optimizations employed to reduce the data motion and cache coherency protocol penalties of the MIC are expected to be of value for CFD and many other unstructured applications as many-core architecture evolves. We explore large-scale distributed-shared memory performance on the Cray XC40 supercomputer, to demonstrate that optimizations employed on Phi hybridize to this context, where each of thousands of nodes are comprised of two sockets of Intel Xeon Haswell CPUs with 32 cores per node.     

Compressing three-dimensional sparse arrays using inter- and intra-task parallelization strategies on Intel Xeon and Xeon Phi

Array operations are useful in a lot of scientific codes. In recent years, several applications, such as the geological analysis and the medical images processing, are processed using array operations for three-dimensional (abbreviate to “3D”) sparse arrays. Due to the huge computation time, it is necessary to compress 3D sparse arrays and use parallel computing technologies to speed up sparse array operations. How to compress the sparse arrays efficiently is an important task for practical applications. Hence, in this paper, two strategies, inter- and intra-task parallelization (abbreviate to “ETP” and “RTP”), are presented to compress 3D sparse arrays, respectively. Each strategy was designed and implemented on Intel Xeon and Xeon Phi, respectively. From experimental results, the ETP strategy achieves 17.5\(\times \) and 18.2\(\times \) speedup ratios based on Intel Xeon E5-2670 v2 and Intel Xeon Phi SE10X, respectively; 4.5\(\times \) and 4.5\(\times \) speedup ratios for the RTP strategy based on these two environments, respectively.     

Fast SIMDized Kalman filter based track fit

Modern high energy physics experiments have to process terabytes of input data produced in particle collisions. The core of many data reconstruction algorithms in high energy physics is the Kalman filter. Therefore, the speed of Kalman filter based algorithms is of crucial importance in on-line data processing. This is especially true for the combinatorial track finding stage where the Kalman filter based track fit is used very intensively. Therefore, developing fast reconstruction algorithms, which use maximum available power of processors, is important, in particular for the initial selection of events which carry signals of interesting physics.One of such powerful feature supported by almost all up-to-date PC processors is a SIMD instruction set, which allows packing several data items in one register and to operate on all of them, thus achieving more operations per clock cycle. The novel Cell processor extends the parallelization further by combining a general-purpose PowerPC processor core with eight streamlined coprocessing elements which greatly accelerate vector processing applications.In the investigation described here, after a significant memory optimization and a comprehensive numerical analysis, the Kalman filter based track fitting algorithm of the CBM experiment has been vectorized using inline operator overloading. Thus the algorithm continues to be flexible with respect to any CPU family used for data reconstruction.Because of all these changes the SIMDized Kalman filter based track fitting algorithm takes 1 μs per track that is 10000 times faster than the initial version. Porting the algorithm to a Cell Blade computer gives another factor of 10 of the speedup.Finally, we compare performance of the tracking algorithm running on three different CPU architectures: Intel Xeon, AMD Opteron and Cell Broadband Engine.     

基于Intel Xeon Phi的激光等离子体粒子模拟研究

激光等离子体粒子模拟广泛用于探索极端物质状态下的科学问题。将一种基于粒子云网格方法的三维等离子体粒子模拟程序LARED P移植到Intel Xeon Phi协处理器上。在移植的过程中,综合运用了Native和Offload两种编程模式：首先运用Native模式对LARED P程序中热点计算任务进行优化研究,通过采用SIMD扩展指令使该计算任务获得了4.61倍的加速;然后运用Offload模式将程序移植到CPU-Intel Xeon Phi异构系统上,并通过使用异步数据传输和双缓冲技术分别提升了程序性能9.8%和21.8%。     

Characterizing and optimizing Java-based HPC applications on Intel many-core architecture

The increasing demand for performance has stimulated the wide adoption of many-core accelerators like Intel^® Xeon Phi^TM Coprocessor, which is based on Intel’s Many Integrated Core architecture. While many HPC applications running in native mode have been tuned to run efficiently on Xeon Phi, it is still unclear how a managed runtime like JVM performs on such an architecture. In this paper, we present the first measurement study of a set of Java HPC applications on Xeon Phi under JVM. One key obstacle to the study is that there is currently little support of Java for Xeon Phi. This paper presents the result based on the first porting of OpenJDK platform to Xeon Phi, in which the HotSpot virtual machine acts as the kernel execution engine. The main difficulty includes the incompatibility between Xeon Phi ISA and the assembly library of Hotspot VM. By evaluating the multithreaded Java Grande benchmark suite and our ported Java Phoenix benchmarks, we quantitatively study the performance and scalability issues of JVM on Xeon Phi and draw several conclusions from the study. To fully utilize the vector computing capability and hide the significant memory access latency on the coprocessor, we present a semi-automatic vectorization scheme and software prefetching model in HotSpot. Together with 60 physical cores and tuning, our optimized JVM achieves averagely 2.7x and 3.5x speedup compared to Xeon CPU processor by using vectorization and prefetching accordingly. Our study also indicates that it is viable and potentially performance-beneficial to run applications written for such a managed runtime like JVM on Xeon Phi.     

申威26010众核处理器上一维FFT实现与优化

根据申威26010众核处理器的特点提出了基于两层分解的一维FFT众核并行算法.该算法基于迭代的Stockham FFT计算框架和Cooley-Tukey FFT算法,将大规模FFT分解成一系列的小规模FFT来计算,并通过设计合理的任务划分方式、寄存器通信、双缓冲以及SIMD向量化等与计算平台相关的优化方法来提高FFT的计算性能.最后对所提出算法的性能进行了测试,相比于单主核上运行的FFTW3.3.4库,获得了平均44.53x的加速比,最高加速比可达56.33x,且其带宽利用率最高可达83.45%.     

基于申威众核处理器的混合并行遗传算法

传统遗传算法求解计算密集型任务时,适应度函数的执行时间增加相当快,致使当种群规模或者进化代数增大时,算法的收敛速度非常缓慢。基于此,设计了"粗粒度-主从式"混合式并行遗传算法（HBPGA）,并在目前TOP500上排名第一的超级计算机神威"太湖之光"平台上实现。该算法模型采用两级并行架构,结合了MPI和Athread两种编程模型,与传统在单核或者一级并行构架的多核集群上实现的遗传算法相比,在申威众核处理器上实现了二级并行,并得到了更好的性能和更高的加速比。实验中,当从核数为16×64时,最大加速比达到544,从核加速比超过31。     

Performance characterization of multi-thread and multi-core processors based XML application oriented networking systems

There is a growing trend to insert application intelligence into network devices. Processors in this type of Application Oriented Networking (AON) devices are required to handle both packet-level network I/O intensive operations as well as XML message-level CPU intensive operations. In this paper, we investigate the performance effect of symmetric multi-processing (SMP) via (1) hardware multi-threading, (2) uni-processor to dual-processor architectures, and (3) single to dual and quad core processing, on both packet-level and XML message-level traffic. We use AON systems based on Intel Xeon processors with hyperthreading, Pentium M based dual-core processors, and Intel’s dual quad-core Xeon E5335 processors. We analyze and cross-examine the SMP effect from both highlevel performance as well as processor microarchitectural perspectives. The evaluation results will not only provide insight to microprocessor designers, but also help system architects of AON types of device to select the right processors.     

Benchmarking Performance of a Hybrid Intel Xeon/Xeon Phi System for Parallel Computation of Similarity Measures Between Large Vectors

The paper deals with parallelization of computing similarity measures between large vectors. Such computations are important components within many applications and consequently are of high importance. Rather than focusing on optimization of the algorithm itself, assuming specific measures, the paper assumes a general scheme for finding similarity measures for all pairs of vectors and investigates optimizations for scalability in a hybrid Intel Xeon/Xeon Phi system. Hybrid systems including multicore CPUs and many-core compute devices such as Intel Xeon Phi allow parallelization of such computations using vectorization but require proper load balancing and optimization techniques. The proposed implementation uses C/OpenMP with the offload mode to Xeon Phi cards. Several results are presented: execution times for various partitioning parameters such as batch sizes of vectors being compared, impact of dynamic adjustment of batch size, overlapping computations and communication. Execution times for comparison of all pairs of vectors are presented as well as those for which similarity measures account for a predefined threshold. The latter makes load balancing more difficult and is used as a benchmark for the proposed optimizations. Results are presented for the native mode on an Intel Xeon Phi, CPU only and the CPU \(+\) offload mode for a hybrid system with 2 Intel Xeons with 20 physical cores and 40 logical processors and 2 Intel Xeon Phis with a total of 120 physical cores and 480 logical processors.     

申威众核处理器的并行NSGA-II算法

非支配排序遗传算法（NSGA-II）在多目标优化领域有着广泛的应用,但在处理复杂问题时运行时间相当长。并行化是提高算法执行速度的有效途径。众核处理器的出现,为实现高度并行奠定了物质基础。基于国产超算“神威·太湖之光”的申威众核处理器平台设计了并行NSGA-II算法（PNSGA-II）,实现了算法基于主核的一级并行和基于主/从核的二级并行。在典型测试函数集上的实验表明,在不影响解的质量前提下,PNSGA-II算法不仅大大加快了执行速度,同时算法的收敛速度也更快。     

面向国产申威26010众核处理器的SpMV实现与优化

世界首台峰值性能超过100P的超级计算机——神威太湖之光已经研制完成,该超级计算机采用了国产申威异构众核处理器,该处理器不同于现有的纯CPU,CPU-MIC,CPU-GPU架构,采用了主-从核架构,单处理器峰值计算能力为3TFlops/s,访存带宽为130GB/s.稀疏矩阵向量乘SpMV（sparse matrix-vector multiplication）是科学与工程计算中的一个非常重要的核心函数,众所周知,其是带宽受限型的,且存在间接访存操作.国产申威处理器给稀疏矩阵向量乘的高效实现带来了很大的挑战.针对申威处理器提出了一种CSR格式SpMV操作的通用异构众核并行算法,该算法从任务划分、LDM空间划分方面进行精细设计,提出了一套动静态buffer的缓存机制以提升向量x的访存命中率,提出了一套动静态的任务调度方法以实现负载均衡.另外还分析了该算法中影响SpMV性能的几个关键因素,并开展了自适应优化,进一步提升了性能.采用Matrix Market矩阵集中具有代表性的16个稀疏矩阵进行了测试,相比主核版最高有10倍左右的加速,平均加速比为6.51.通过采用主核版CSR格式SpMV的访存量进行分析,测试矩阵最高可达该处理器实测带宽的86%,平均可达到47%.     

A mixed-precision algorithm for the solution of Lyapunov equations on hybrid CPU-GPU platforms

We describe a hybrid Lyapunov solver based on the matrix sign function, where the intensive parts of the computation are accelerated using a graphics processor (GPU) while executing the remaining operations on a general-purpose multi-core processor (CPU). The initial stage of the iteration operates in single-precision arithmetic, returning a low-rank factor of an approximate solution. As the main computation in this stage consists of explicit matrix inversions, we propose a hybrid implementation of Gauß-Jordan elimination using look-ahead to overlap computations on GPU and CPU. To improve the approximate solution, we introduce an iterative refinement procedure that allows to cheaply recover full double-precision accuracy. In contrast to earlier approaches to iterative refinement for Lyapunov equations, this approach retains the low-rank factorization structure of the approximate solution. The combination of the two stages results in a mixed-precision algorithm, that exploits the capabilities of both general-purpose CPUs and many-core GPUs and overlaps critical computations. Numerical experiments using real-world data and a platform equipped with two Intel Xeon QuadCore processors and an Nvidia Tesla C1060 show a significant efficiency gain of the hybrid method compared to a classical CPU implementation.     

Efficient and retargetable SIMD translation in a dynamic binary translator

The single‐instruction multiple‐data (SIMD) computing capability of modern processors is continually improved to deliver ever better performance and power efficiency. For example, Intel has increased SIMD register lengths from 128 bits in streaming SIMD extension to 512 bits in AVX‐512. The ARM scalable vector extension supports SIMD register length up to 2048 bits and includes predicated instructions. However, SIMD instruction translation in dynamic binary translation has not received similar attention. For example, the widely used QEMU emulates guest SIMD instructions with a sequence of scalar instructions, even when the host machines have relevant SIMD instructions. This leaves significant potential for performance enhancement. We propose a newly designed SIMD translation framework for dynamic binary translation, which takes advantage of the host's SIMD capabilities. The proposed framework has been built in HQEMU, an enhanced QEMU with a separate thread for applying LLVM optimizations. The current prototype supports ARMv7, ARMv8, and IA32 guests on the X86‐64 AVX‐2 host. Compared with the scalar‐translation version HQEMU, our framework runs up to 1.84 times faster on Standard Performance Evaluation Corporation 2006 CFP benchmarks and up to 6.81 times faster on selected real applications.     

Using the Xeon Phi Platform to Run Speculatively-Parallelized Codes

Intel Xeon Phi accelerators are one of the newest devices used in the field of parallel computing. However, there are comparatively few studies concerning their performance when using most of the existing parallelization techniques. One of them is thread-level speculation, a technique that optimistically tries to extract parallelism of loops without the need of a compile-time analysis that guarantees that the loop can be executed in parallel. In this article we evaluate the performance delivered by an Intel Xeon Phi coprocessor when using a software, state-of-the-art thread-level speculative parallelization library in the execution of well-known benchmarks. We describe both the internal characteristics of the Xeon Phi platform and the particularities of the thread-level speculation library being used as benchmark. Our results show that, although the Xeon Phi delivers a relatively good speedup in comparison with a shared-memory architecture in terms of scalability, the relatively low computing power of its computational units when specific vectorization and SIMD instructions are not fully exploited makes this first generation of Xeon Phi architectures not competitive (in terms of absolute performance) with respect to conventional multicore systems for the execution of speculatively parallelized code.     

Numerical reproducibility for the parallel reduction on multi- and many-core architectures

On modern multi-core, many-core, and heterogeneous architectures, floating-point computations, especially reductions, may become non-deterministic and, therefore, non-reproducible mainly due to the non-associativity of floating-point operations. We introduce an approach to compute the correctly rounded sums of large floating-point vectors accurately and efficiently, achieving deterministic results by construction. Our multi-level algorithm consists of two main stages: first, a filtering stage that relies on fast vectorized floating-point expansion; second, an accumulation stage based on superaccumulators in a high-radix carry-save representation. We present implementations on recent Intel desktop and server processors, Intel Xeon Phi co-processors, and both AMD and NVIDIA GPUs. We show that numerical reproducibility and bit-perfect accuracy can be achieved at no additional cost for large sums that have dynamic ranges of up to 90 orders of magnitude by leveraging arithmetic units that are left underused by standard reduction algorithms.     

Xeon Phi平台上基于模板优化的3D-GVF场计算加速

3D梯度向量流场（3D GVF field）广泛应用于多种3D图像分析算法中,其计算需要多次迭代,计算量大,如何提高其计算速度具有重要的研究意义。面向Intel Xeon Phi众核集成架构,首次进行了3D GVF场计算的加速优化。首先,挖掘3D图像像素点间存在的天然并行性,发挥众核架构优势,尝试线程级并行（多核）和数据级并行（SIMD）。其次,3D GVF场的计算过程是一种典型的3D 7点模板运算,结合Xeon Phi架构的L2 缓存规格,提出一种高效的数据分块策略,充分挖掘数据的时/空局部性,有效缓解模板计算引起的缓存缺失,提升了计算性能。实验结果表明,引入模板优化技术能显著提升3D GVF场的计算速度,在图像维度为5123时,所提方法在57核Xeon Phi平台上的性能相比在2.6GHz 8核16线程的Intel Xeon E5 2670 CPU上的性能,加速比可达2.77。     

Explicit Fourth-Order Runge–Kutta Method on Intel Xeon Phi Coprocessor

This paper concerns an Intel Xeon Phi implementation of the explicit fourth-order Runge–Kutta method (RK4) for very sparse matrices with very short rows. Such matrices arise during Markovian modeling of computer and telecommunication networks. In this work an implementation based on Intel Math Kernel Library (Intel MKL) routines and the authors’ own implementation, both using the CSR storage scheme and working on Intel Xeon Phi, were investigated. The implementation based on the Intel MKL library uses the high-performance BLAS and Sparse BLAS routines. In our application we focus on OpenMP style programming. We implement SpMV operation and vector addition using the basic optimizing techniques and the vectorization. We evaluate our approach in native and offload modes for various number of cores and thread allocation affinities. Both implementations (based on Intel MKL and made by the authors) were compared in respect of the time, the speedup and the performance. The numerical experiments on Intel Xeon Phi show that the performance of authors’ implementation is very promising and gives a gain of up to two times compared to the multithreaded implementation (based on Intel MKL) running on CPU (Intel Xeon processor) and even three times in comparison with the application which uses Intel MKL on Intel Xeon Phi.     

Exploiting task and data parallelism in ILUPACK’s preconditioned CG solver on NUMA architectures and many-core accelerators

We present specialized implementations of the preconditioned iterative linear system solver in ILUPACK for Non-Uniform Memory Access (NUMA) platforms and many-core hardware co-processors based on the Intel Xeon Phi and graphics accelerators. For the conventional x86 architectures, our approach exploits task parallelism via the OmpSs runtime as well as a message-passing implementation based on MPI, respectively yielding a dynamic and static schedule of the work to the cores, with different numeric semantics to those of the sequential ILUPACK. For the graphics processor we exploit data parallelism by off-loading the computationally expensive kernels to the accelerator while keeping the numeric semantics of the sequential case.     

Sparse matrix–vector multiplication on the Single-Chip Cloud Computer many-core processor

The microprocessor industry has responded to memory, power and ILP walls by turning to many-core processors, increasing parallelism as the primary method to improve processor performance. These processors are expected to consist of tens or even hundreds of cores. One of these future processors is the 48-core experimental processor Single-Chip Cloud Computer (SCC). The SCC was created by Intel Labs as a platform for many-core software research.