A5102008886- Advanced Data Storage Technologies
- Numerical Methods and Algorithms
- Stochastic processes and financial applications
- Advancements in PLL and VCO Technologies
- Mathematical Approximation and Integration
- Analog and Mixed-Signal Circuit Design
- Chaos-based Image/Signal Encryption
- Network Time Synchronization Technologies
- Financial Markets and Investment Strategies
- Quantum Computing Algorithms and Architecture
- Statistical and numerical algorithms
- Parallel Computing and Optimization Techniques
- Financial Risk and Volatility Modeling
- Probabilistic and Robust Engineering Design
- GNSS positioning and interference
- Advanced Optical Imaging Technologies
- Algorithms and Data Compression
- Genomics and Phylogenetic Studies
- Image Processing Techniques and Applications
- Cellular Automata and Applications
- Inertial Sensor and Navigation
- Distributed and Parallel Computing Systems
- Evolutionary Algorithms and Applications
- Advanced Data Compression Techniques
- Embedded Systems Design Techniques
University of Science and Technology of China
2014-2015
University of Edinburgh
2008-2013
This paper explores the pros and cons of reconfigurable computing in form FPGAs for high performance efficient computing. In particular, presents results a comparative study between three different acceleration technologies, namely, Field Programmable Gate Arrays (FPGAs), Graphics Processor Units (GPUs), IBM’s Cell Broadband Engine (Cell BE), design implementation widely-used Smith-Waterman pairwise sequence alignment algorithm, with general purpose processors as base reference...
Quasi-Monte Carlo simulation is a special Monte method that uses quasi-random or low-discrepancy numbers as random sample sets. In many applications, this has proved advantageous compared to the traditional method, which pseudo-random numbers, thanks its faster convergence and higher level of accuracy. This article presents design implementation massively parallelized engine on an FPGA-based supercomputer, called Maxwell. It also compares with equivalent graphics processing units (GPUs)...
Random number generation is a very important operation in computational science e.g. Monte Carlo simulations methods. It also computationally intensive especially for high quality random generation. In this paper, we present the design and implementation of parallel one most widely used generators, namely Mersenne Twister. The latter performance computing applications such as financial computing. Implementations our Twister generator core on Xilinx Virtex4 FPGAs achieve throughput 26.13...
Monte-Carlo simulation is a very widely used technique in scientific computations general with huge computation benefits solving problems where closed form solutions are impossible to derive. This also characterized by high degree of parallelism as large number different paths need be calculated, which makes it ideal for parallel hardware implementation. paper illustrates the such implementation context financial computing implements engine on an FPGA-based supercomputer, called Maxwell,...
The time-to-digital converter(TDC) is an equipment which aims to measure the accurate time of edges input signal. Our work present I/O Tile based multi-phase clock TDC, implemented in Field-Programmable-Gate-Array(FPGA). A hit signal sampled by 8 equidistant phase-shifted clocks Tile. differential standard connecting buffered buffer and split into two complementary outputs before feeding adjacent ISERDESes. ISERDESes are configured as oversample mode, used capture 2 phase DDR data. One...
The valuation of optimal exercise American-style options is one the most important problems in option pricing theory. Unlike European options, American have feature early exercise, which makes it hard to simulate using simple Monte Carlo method. A number extended methods been published recently; Least-Squares (LSMC) suggested by Longstaff and Schwartz adopted algorithms industry. Although hardware acceleration technique has used financial computing for several years, there not any...
System control and the collection of analog signals are fundamental tasks in many embedded computer systems such as found automotive, communication, sensor network domains. Often latency is critical FPGAs an attractive alternative. Traditionally, external ADC (analog to digital converter) chips have been used for signal conversion transfer FPGA(s). In large-scale quantum system experiments, implementation this classic infrastructure a challenge. particular, FPGA-based must work liquid helium...
As Field Programmable Gate Arrays (FPGAs) get faster and denser, the scope of their applications is getting wider. High performance computing applications, for instance, are an example such application expansion driven by FPGAs' increasing computational power coupled with relatively low consumption compared to state-of-the-art microprocessor technology. However, one major hurdle facing FPGAs in high arena, addition level programming model, efficiency implementing double precision...
Quasi-Monte Carlo simulation is a specialized Monte method which uses quasi-random, or low-discrepancy, numbers as the stochastic parameters. In many applications, this has proved advantageous compared to traditional method, pseudo-random numbers, it converges relatively quickly, and with better level of accuracy. We implemented massively parallelized engine on FPGA-based supercomputer, called Maxwell, developed at University Edinburgh. Maxwell consists 32 IBM Intel Xeon blades each hosting...
In this paper, we present a high performance scalable FPGA design and implementation of an interest rate derivative pricing engine that targets on the cap pricing. The consists Gaussian random number generator, based Mersenne Twister uniform Monte Carlo path generation which calculates prices LIBOR market model. We implemented Maxwell supercomputer using up to 32 Xilinx XC4VFX100 nodes. have also compared our hardware with equivalent optimized pure software running 2.8GHz Xeon processors 1...
An improved anti-aliasing sampling algorithm is submitted to reduce the increasing memory consumption caused by super-sampling in mobile devices. Six-point anisotropy blends two samples of a pixel, as well nearby pixels. Experiment results showed that six-point has reduced 50% than traditional FLIPQUAD algorithm. This method similar quality with only consumption.