Performance Analysis of Existing SIMD Architectures

Due to the high computational complexity of the full-search algorithm for video coding, techniques to accelerate the execution of the full-search algorithm are needed for real-time implementations of current video compression standards. In this paper, a technique of using eight-bit partial sums of four luminance values is proposed to reduce the computational complexity of the original full-search algorithm without loss of its accuracy. Furthermore, since the proposed technique is suitable for the implementation on the single instruction multiple data (SIMD) architectures, the byte-type data-parallelism of the SIMD architecture can be utilized to further accelerate the execution of the proposed technique. Simulation results for the benchmark video test sequences demonstrate that the proposed technique can significantly accelerate the execution of the full-search algorithm on SIMD architectures without loss of its accuracy.

Energy efficiency has become one of the most important topics in computing. To meet the ever increasing demands of the mobile market, the next generation of processors will have to deliver a high compute performance at an extremely limited energy budget. Wide single instruction, multiple data (SIMD) architectures provide a promising solution, as they have the potential to achieve high compute performance at a low energy cost. We propose a configurable wide SIMD architecture that utilizes explicit datapath techniques to further optimize energy efficiency without sacrificing computational performance. To demonstrate the efficiency of the proposed architecture, multiple instantiations of the proposed wide SIMD architecture and its automatic bypassing counterpart, as well as a baseline RISC processor, are implemented. Extensive experimental results show that the proposed architecture is efficient and scalable in terms of area, performance, and energy. In a 128-PE SIMD processor, the proposed architecture is able to achieve an average of 206 times speed-up and reduces the total energy dissipation by 48.3 % on average and up to 94 %, compared to a reduced instruction set computing (RISC) processor. Compared to the corresponding SIMD architecture with automatic bypassing, an average of 64 % of all register file accesses is avoided by the 128-PE, explicitly bypassed SIMD. For total energy dissipation, an average of 27.5 %, and maximum of 43.0 %, reduction is achieved.

The aim of this study is to present the steps of optimization the H.264 / AVC video decoder for MIPS32 DSP ASE SIMD (Single-instruction, multiple-data) processor architecture. In particular, SIMD architecture can be very efficient in the implementation of video applications which have simple operations on 8-bit or 16-bit samples. Besides already mentioned architecture it is also important to get familiar with the H.264 / AVC video code standards. The parts of decoder that were subject of optimization were detected as the biggest consumers of CPU resources. After optimization, overall system speedup gain was 36%.

The use of a wide Single-Instruction-Multiple-Data (SIMD) architecture is a promising approach to build energy-efficient high performance embedded processors. In this paper, based on our design framework for low-power SIMD processors, we propose a multiply-accumulate (MAC) unit with variable number of accumulator registers. The proposed MAC unit exploits both the merits of merged operation and register tiling. A Convolutional Neural Network (CNN) is a popular learning based algorithm due to its flexibility and high accuracy. However, a CNN-based application is often computationally intensive as it applies convolution operations extensively on a large data set. In this work, a CNN-based intelligent learning application is analyzed and mapped in the context of SIMD architectures. Experimental results show that the proposed architecture is efficient. In a 64-PE instance, the proposed SIMD processor with MAC4reg achieves an effective performance of 63.2 GOPS. Compared to the two baseline SIMD processors without MAC4reg, the proposed design brings 54.0% and 32.1% reduction in execution time, and 20.5% and 35.1% reduction in energy consumption respectively.

The general-purpose neural network processor is designed for the most neural networks algorithm and is required for variable bit length data processing ability. This paper proposes a processor that is based on SIMD (Single Instruction Multiple Data) architecture with three data bit mode: 8-bit, 16-bit and 32-bit. It can use the memory and ALU sufficiently when the bit mode changes. The processor is designed basing on 0.25–micron process technology and it can be synthesized at 50MHz with PKS of Cadence Inc. The experiment result shows that the processor can implement the neural network in highly parallel.KeywordsLogic ModuleExtreme Learn MachineHardware ImplementationRegister FileTime SpendingThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This paper presents an efficient, fine-grained parallel algorithm for solving the Eikonal equation on triangular meshes. The Eikonal equation, and the broader class of Hamilton-Jacobi equations to which it belongs, have a wide range of applications from geometric optics and seismology to biological modeling and analysis of geometry and images. The ability to solve such equations accurately and efficiently provides new capabilities for exploring and visualizing parameter spaces and for solving inverse problems that rely on such equations in the forward model. Efficient solvers on state-of-the-art, parallel architectures require new algorithms that are not, in many cases, optimal, but are better suited to synchronous updates of the solution. In previous work [W. K. Jeong and R. T. Whitaker, SIAM J. Sci. Comput., 30 (2008), pp. 2512-2534], the authors proposed the fast iterative method (FIM) to efficiently solve the Eikonal equation on regular grids. In this paper we extend the fast iterative method to solve Eikonal equations efficiently on triangulated domains on the CPU and on parallel architectures, including graphics processors. We propose a new local update scheme that provides solutions of first-order accuracy for both architectures. We also propose a novel triangle-based update scheme and its corresponding data structure for efficient irregular data mapping to parallel single-instruction multiple-data (SIMD) processors. We provide detailed descriptions of the implementations on a single CPU, a multicore CPU with shared memory, and SIMD architectures with comparative results against state-of-the-art Eikonal solvers.

Existing fault simulation techniques have a worst-case computational complexity of O(n**2), where n is the number of devices in the circuit. A parallel hardware accelerated fault simulator (PHAFS) has been proposed in order to reduce the complexity to O(L**2), where L is the number of levels of switches encountered when traversing from output to input. An overview of the PHAFS architecture and system is presented. The PHAFS hardware component uses a massively parallel SIMD (single instruction, multiple data) architecture based on an application specific integrated circuit to perform two-dimensional parallel switch level fault simulation up to 32 switches and 192 faults. >

In the past two years the processing power of video graphics cards has quadrupled and is approaching super computer levels. State-of-the-art graphical processing units (GPU) boast of theoretical computational performance in the range of 1.5 trillion floating point operations per second (1.5 Teraflops). This processing power is readily accessible to the scientific community at a relatively small cost. High level programming languages are now available that give access to the internal architecture of the graphics card allowing greater algorithm optimization. This research takes memory access expensive portions of an image-based iris identification algorithm and hosts it on a GPU using the C++ compatible CUDA language. The selected segmentation algorithm uses basic image processing techniques such as image inversion, value squaring, thresholding, dilation, erosion and memory/computationally intensive calculations such as the circular Hough transform. Portions of the iris segmentation algorithm were accelerated by a factor of 77 over the 2008 GPU results. Some parts of the algorithm ran at speeds that were over 1600 times faster than their CPU counterparts. Strengths and limitations of the GPU Single Instruction Multiple Data architecture are discussed. Memory access times, instruction execution times, programming details and code samples are presented as part of the research.

Software defined radio (SDR) systems require programmable architectures that provide high performance combined with high energy efficiency. Wide single instruction, multiple data (SIMD) architectures could potentially satisfy these demands, but only few publications examine the interdependence between the SIMD width and the performance of typical SDR algorithms. In this paper, we present a detailed analysis of three SDR algorithm classes on a newly developed scalable SIMD architecture, which combines SIMD processing with long instruction word (LIW) execution. Although radix-2 and mixed-radix fast Fourier transform, sphere decoding, and low-density parity-check decoding all benefit from SIMD processing, our results show that there are different constraints on the performance.

With the multitude of existing and upcoming wireless standards, it is becoming increasingly difficult for hardware-only baseband processing solutions to adapt to the rapidly changing wireless communication landscape. Software defined radio (SDR) promises to deliver a cost effective and flexible solution by implementing a wide variety of wireless protocols in software. In previous work, a fully programmable multicore architecture, SODA, was proposed that was able to meet the real-time requirements of 3G wireless protocols. SODA consists of one ARM control processor and four wide single instruction multiple data (SIMD) processing elements. Each processing element consists of a scalar and a wide 512-bit 32-lane SIMD datapath. A commercial prototype based on the SODA architecture, Ardbeg (named after a brand of Scotch whisky), has been developed. In this paper, we present the architectural evolution of going from a research design to a commercial prototype, including the goals, tradeoffs, and final design choices. Ardbegpsilas redesign process can be grouped into the following three major areas: optimizing the wide SIMD datapath, providing long instruction word (LIW) support for SIMD operations, and adding application-specific hardware accelerators. Because SODA was originally designed with 180 nm technology, the wide SIMD datapath is re-optimized in Ardbeg for 90 nm technology. This includes re-evaluating the most efficient SIMD width, designing a wider SIMD shuffle network, and implementing faster SIMD arithmetic units. Ardbeg also provides modest LIW support by allowing two SIMD operations to issue in the same cycle. This LIW execution supports SDR algorithmspsila most common parallel SIMD execution patterns with minimal hardware overhead. A viable commercial SDR solution must be competitive with existing ASIC solutions. Therefore, algorithm-specific hardware is added for performance bottleneck algorithms while still maintaining enough flexibility to support multiple wireless protocols. The combination of these architectural improvements allows Ardbeg to achieve 1.5-7x speedup over SODA across multiple wireless algorithms while consuming less power.

Use of single instruction multiple data (SIMD) functional units enables multimedia systems to exploit parallelism to a higher degree resulting in significant system performance improvements. While implementation of whole SIMD system functionality for an application results in wastage of area resources, we have observed that for a specific multimedia application, we only need to implement a customized SIMD unit that is a subset of whole SIMD standard implementation. Based on this study, we have proposed an extension to the traditional system design and synthesis flow by integrating a methodology of SIMD unit synthesis. Our system synthesizes a customized SIMD unit along with an extended instruction set and generates an equivalent version of assembly code for the application using the extended instruction set. The results of area and performance obtained by experimenting over our implementation of AltiVec compatible customized SIMD units show the effectiveness of our approach.

SIMD (Single Instruction Multiple Data) extensions are acceleration components integrated in general processor, aiming at extracting instruction and data level parallelism of multimedia and scientific calculation programs. Currently, most of the automatic vectorization methods for SIMD architectures are based on innermost loops. Inner loop vectorization is the common approach for auto-vectorization. This method has been used for many years and its efficiency is widely accepted by people. In this paper, we put forward a better method than inner loop vectorization for some loop nests, which is outer loop vectorization. Outer loop vectorization method means vectorizing the outer loop directly. It can extract more data level parallelism and make the most use of the spatial locality to improve the program efficiency than the inner one which is more suitable for some loop nests. This paper presents the realization. In this paper, we first revisit the preliminary analysis of outer loop vectorization based on Open64 complier. And then, data type conversion and code generation is presented in detail. Finally, we propose two optimization methods, capable of boosting the performance of outer loop vectorization to achieve the acceleration of 20% on average, 50% at most.

Recent microprocessors have included SIMD (single instruction multiple data) extensions into their instruction set architecture to improve the performance of multimedia applications. SIMD instructions speed up the execution of programs but pose lots of challenges to software developers. An efficient matrix-based splitter (or merger), which can split an N × N 2-D DCT block into four N/2 × N/2 or two N × N/2 (or N/2 × N) 2-D DCT blocks (or merger small size blocks into a large size one), specialized for SIMD architectures is presented in this paper. The programming-level complexity of the proposed methods is lower than that of the direct approach. Furthermore, even without using SIMD instructions, the algorithmic-level complexity of the proposed DCT splitter/merger is still lower than that of the direct one and is the same as that of the most efficient approach existed in the literature. When N = 8, our method can be applied to act as a transcoder between the latest video coding standards AVC/H.264 and the older ones, such as MPEG-1, MPEG-2 and MPEG-4 part 2. We also provide the image quality tests to show the performance of the proposed 2-D DCT splitter and merger.

In-memory computing (IMC) has been proposed to overcome the von Neumann bottleneck in data-intensive applications. However, existing IMC solutions could not achieve both high parallelism and high flexibility, which limits their application in more general scenarios: As a highly parallel IMC design, the functionality of a MAC crossbar is limited to the matrix-vector multiplication; Another IMC method of logic-in-memory (LiM) is more flexible in supporting different logic functions, but has low parallelism. To improve the LiM parallelism, we are inspired by investigating how the single-instruction, multiple-data (SIMD) instruction set in conventional CPU could potentially help to expand the number of LiM operands in one cycle. The biggest challenge is the inefficiency in handling non-continuous data in parallel due to the SIMD limitation of (i) continuous address, (ii) limited cache bandwidth, and (iii) large full-resolution parallel computing overheads. This article presents GRAPHIC, the first reported in-memory SIMD architecture that solves the parallelism and irregular data access challenges in applying SIMD to LiM. GRAPHIC exploits content-addressable memory (CAM) and row-wise-accessible SRAM. By providing the in-situ, full-parallelism, and low-overhead operations of address search, cache read-compute-and-update, GRAPHIC accomplishes high-efficiency gather and aggregation with high parallelism, high energy efficiency, low latency, and low area overheads. Experiments in both continuous data access and irregular data pattern applications show an average speedup of 5x over iso-area AVX-like LiM, and 3-5x over the emerging CAM-based accelerators of CAPE and GaaS-X in advanced techniques.

Several applications have problems with the representation of the real numbers because of its drawbacks like the propagation and the accumulation of errors. These numbers have a fixed length format representation that provides a large dynamic range, but on the other hand it causes truncation of some parts of the numbers in case of a number that needs to be represented by a long stream of bits. Researchers suggested many solutions for these errors, one of these solutions is the Multi-Number (MN) system. MN system represents the real number as a vector of floating-point numbers with controlled accuracy by adjusting the length of the vector to accumulate the non-overlapping real number sequences. MN system main drawback is the MN computations that are iterative and time consuming, making it unsuitable for real time applications. In this work, the Single Instruction Multiple Data (SIMD) model supported in modern CPUs is exploited to accelerate the MN Computations. The basic arithmetic operation algorithms had been adjusted to make use of the SIMD architecture and support both single and double precision operations. The new architecture maintains the same accuracy of the original one, when was implemented for both single and double precision. Also, in this paper the normal Gaussian Jordan Elimination algorithm was proposed and used to get the inverse of the Hilbert Matrix, as an example of ill-conditioned matrices, instead of using iterative and time-consuming methods. The accuracy of the operations was proved by getting the inverse of the Hilbert Matrix and verify that the multiplication of the inverse and the original matrix producing the unity matrix. Hilbert Matrix inverse execution time was accelerated and achieved a speedup 3x, compared to the original NM operations. In addition to the previous, the accelerated MN system version was used to solve the polynomial regression problem.