The facility allocation of the supply chain is critical since it directly influences cost efficiency, customer service, supply chain responsiveness, risk reduction, network optimization, and overall competitiveness. When enterprises deploy their facilities wisely, they may achieve operational excellence, exceed customer expectations, and obtain a competitive advantage in today's volatile business climate. Due to this reason, a multi-objective facility allocation problem is introduced in this research with cooperative-based multi-level backup coverage considering distance-based facility attractiveness. The facility of the coverage is further described as two different layers of the coverage process, where demand can be covered as full, partial, and no coverage by their respective facilities. The main objectives of this facility allocation problem are to maximize the coverage of the facility to maximize overall facility coverage in the supply chain network and simultaneously minimize the overall cost.

Mammograms are always used to detect signs of breast cancer. Texture-analysis techniques were applied to determine imaging biomarkers consisting of mean, contrast, correlation, energy and homogeneity features of parametric maps, and they can be utilized to retravel specific features of symptoms from medical images produced by X-ray, magnetic resonance imaging, computed tomography, and so forth. Gray level run-length matrix is the one of the texture extraction methods which has been successfully used to facilitate medical image analysis. However, it is computation-intensive method. We implemented it on GPU to accelerating extraction process for mammograms. The proposed method achieves significant speedup over CPU-based implementation.

Seam carving is a content-aware image processing approach that has been successfully applied to resizing or removing objects from digital images. Seam-carved images are hard to identify from the original image, making this detection method an important and attractive research topic. Existing methods are based on steganography attacks or statistical features, and their trained and tested models are all constructed using support vector machines (SVMs). The trained models of these methods take a long time to build from a limited number of images. This paper presents a new patch-based sobel operator (PSO) method using SVM and extreme learning machines (ELMs) based on the patch analysis method with square-based features. The ELM adopts five types of neurons to obtain five prediction results, which are then combined using the jury voting scheme in order to improve the accuracy. In addition, combinatorial fusion is first used to choose two of five prediction results according to the diversity rank/score graph, and these two results are then combined to make the final decision. The PSO method in the ELM with jury voting and combinatorial fusion achieves accuracies of 70.60-98.80% and 95.58-99.36% for 10-50% seam-carved and seam-insertion images, respectively. Additionally, for the PSO method, the ELM required only 1.2% of the training time required by the SVM. In conclusion, the PSO method in the ELM is useful for constructing trained models from a large number of images.

The increasing demand for machine learning computation contributes to the convergence of high-performance computing and cloud computing, in which the virtualization of Graphics Processing Units (GPUs) becomes a critical issue. Although many GPGPU virtualization frameworks have been proposed, their performance is limited by the bandwidth of data transactions between the virtual machine (VM) and host. In this paper, we present a virtualization framework, qCUDA, to improve the performance of compute unified device architecture (CUDA) programs. qCUDA is based on the virtio framework, providing the para-virtualized driver and the device module for performing the interaction with the API remoting and memory management methods. In our test environment, qCUDA can achieve above 95% of the bandwidth efficiency for most results by comparing it with the native. Also, qCUDA has the features of flexibility and interposition. It can execute CUDA-compatible programs in the Linux and Windows VMs, respectively, on QEMU-KVM hypervisor for GPGPU virtualization.

A phylogenetic tree is used to present the evolutionary relationships among the interesting biological species based on the similarities in their genetic sequences. The UPGMA is one of the popular algorithms to construct a phylogenetic tree according to the distance matrix created by the pairwise distances among taxa. To solve the performance issue of the UPGMA, the implementation of the UPGMA method on a single GPU has been proposed. However, it is not capable of handling the large taxa set. This work describes a novel parallel UPGMA approach on multiple GPUs that is able to build a tree from extremely large datasets. The experimental results show that the proposed approach with 4 NVIDIA GTX 980 achieves an approximately × fold speedup over the implementation of UPGMA on CPU and GPU, respectively.

High-end graphics processing units (GPUs), such as NVIDIA Fermi/Tesla series cards, are widely applied to the high performance computing fields in a decade. NVIDIA releases Tegra K1, called Jetson TK1, which contains 4 ARM Cortex-A15 CPUs and 192 CUDA cores (Kepler GPU) is an embedded board with low cost, low power consumption, and high applicability advantages for several specific applications. In the previous work, we have constructed a bioinformatics platform based on single NVIDIA Jetson TK1, and ClustalWtk tool for multiple sequence alignments was designed on this platform. With more and more biological sequences generated and high computing power requirement, it is necessary to construct a cluster platform based on multiple NVIDIA Jetson TK1. In this paper, a bioinformatics platform based on a cluster of multiple NVIDIA Jetson TK1 is constructed. Besides, two job assignment modes are proposed in this platform: one is a “user selected” mode and another is an “automatic assigned” mode. ClustalWtk tool also is ported on this platform with these two modes, respectively.

Summary Wafer processing technology has been improving rapidly. Moore's law has been exceeded as the number of transistors in a dense integrated circuit, now increases threefold or more, approximately every year. The integrated circuit has gone from very large scale to giga large scale. The extraction of subcircuits has therefore become computation‐intensive. In this paper, we propose an efficient bit‐parallel subcircuit extraction algorithm using graphic processing units. We conducted experimental trials and demonstrated that the proposed algorithm can achieve high throughput, suggesting practical applications in the extraction of subcircuits. Copyright © 2015 John Wiley & Sons, Ltd.

GPU computing has gradually become the mainstream to do high-speed computing fields, such as the meteorology, image and video processing, fluid dynamics simulation, seismic analysis, and etc. How to efficiently port an existing program on CPU to its CUDA program on GPU is an important issue. From the previous works, the porting approach can be generalized and classified into two categories: Rewrite Parallel Algorithm (abbreviate to RPA) and Modify Original Library (abbreviate to MOL). For the RPA, the programmers need to understand the original sequential or parallel algorithm on CPU absolutely and then write the CUDA program on GPU directly. For the MOL, the programmers need to analyze (profile) the existing program on CPU at first to find the most spend time libraries (or functions), then they are modified greatly (rewritten in general) to become CUDA programs (kernel functions). There are several disadvantages for the RPA and MOL, especially for the porting time and executing results. Hence, in this paper, a new approach, called innovative systematic contract (abbreviate to ISC), is proposed to allow programmers to port an existing CPU program to its CUDA program by modifying the libraries lightly. The program, BLASTN v2.2.27, was ported into a CUDA version, called CUDA-BLASTN v1, by the ISC. From the experimental results, by comparing with BLASTN v2.2.27, CUDA-BLASTN v1 achieves 5x speedup ratio and obtains almost the same executing results.

Summary Constructing phylogenetic trees is of priority concern in computational biology, especially for developing biological taxonomies. As a conventional means of constructing phylogenetic trees, unweighted pair group method with arithmetic (UPGMA) is also an extensively adopted heuristic algorithm for constructing ultrametric trees (UT). Although the UT constructed by UPGMA is often not a true tree unless the molecular clock assumption holds, UT is still useful for the clocklike data. Moreover, UT has been successfully adopted in other problems, including orthologous‐domain classification and multiple sequence alignment. However, previous implementations of the UPGMA method have a limited ability to handle large taxa sets efficiently. This work describes a novel graphics processing unit (GPU)‐UPGMA approach, capable of providing rapid construction of extremely large datasets for biologists. Experimental results indicate that the proposed GPU‐UPGMA approach achieves an approximately 95× speedup ratio on NVIDIA Tesla C2050 GPU over the implementation with 2.13 GHz CPU. The developed techniques in GPU‐UPGMA also can be applied to solve the classification problem for large data set with more than tens of thousands items in the future.Copyright © 2014 John Wiley & Sons, Ltd.

Compound comparison is an important task for computational chemistry. By the comparison reulsts, potential inhibitors can be found and then used for the following experiments. The time complexity of a pairwise compound comparison is O(n2), where n is the maximal length of compounds. In general, the compound length is small, and the cost of computation time is short. However, more and more compounds have been synthesized and extracted now, even more than ten of millions. Therefore, it still will be time-consuming when comparing with a large amount of compounds (multiple compound comparisons). In this paper, we propose a parallel algorithm for multiple compound comparisons on multi-GPUs. Four load-balancing strategies were considered in the proposed algorithm in order to accelerate the computation speed among thread blocks on GPUs. The proposed algorithm was implemented by C+OpenMP+CUDA, and achieved more than 50 times speedup by comparing with its CPU version under the experiemtal results.

Nowadays, with the explosive growth of the network technologies many new applications and services have been developed on Internet. World Wide Web can provide these services provided without the limitation of time and location. Obviously, the number of user is dramatically increasing from amount of the visitations of web pages. In our previous work, we proposed an algorithm to discover more significant information from visited web pages to provide this information to web designers or policy makers to adjust the presentation of their Web contents. However, this algorithm is time-consuming approach due to it needs to scan the whole database many times. Therefore, we propose a GPGPU-based Preference Utility algorithm to enhance the performance by GPGPU parallel model. The proposed algorithm is developed on NVIDIA CUDA architecture. The experimental results show that the proposed method can achieve about 7x times over CPU-based method. The proposed algorithm can used to mine the information from web log data efficiently.

Cloud computing, an emerging Internet-based development provides various platform and software services has become a significant issue. Many bioinformatics tools developed based on the Internet for biologists without reconstructing the whole software in the local system. Therefore, bioinformatics as a service is a new significant demand of cloud computing that integrates the bioinformatics tools to cloud platform to provide more efficient bio-computing services. In this paper, we propose a virus analysis service on the cloud. Norovirus is used as case study for this service. The result demonstrates that the proposed cloud service is able to be easily used to analyze virus data by using remote computing resources.

The Smith-Waterman algorithm is the most widely used algorithm to analyze the similarity between protein and DNA sequences and suitable for the database search due to its high sensitivity. However, Smith-Waterman still is a very time-consuming method. CUDA programming can efficiently improve the computations by using the computing power of the massive computing hardware as GPUs. In this paper, we proposed an efficient frequency based filter method instead of just speed up the Smith-Waterman comparison but waste computing resource to deal with those unnecessary comparisons. We implemented the Smith-Waterman algorithm by introduction of the techniques from earlier researches and add in our real-time filter method on Graphic Processing Units to filter unnecessary comparisons. We also design a user friendly interface to provide the service in the potential clouding computing environment. In our research we choose two data sets, H1N1 VH protein database and Human protein database then compare CUDA-SW and CUDA-SW with filter, we called CUDA-SWf we can obtain up to 41% performance improve from reduce unnecessary sequence alignments.

Multiple sequence alignments with constrains has become an important problem in the computational biology. The concept of constrained sequence alignment is proposed to incorporate the biologist's domain knowledge into sequence alignments such that the user-specified residues/segments are aligned together in the alignment results. Over the past decade, a series of constrained multiple sequence alignment tools were proposed in the literature. GPU-REMuSiC is a newest tool with the regular expression constrains and uses the graphics processing units (GPUs) with CUDA. GPU-REMuSiC can achieve 29× speedups for overall computation time by the experimental results. However, the execution environment of GPU-REMuSiC need to build, and it's a threshold for biologists to set up. Therefore, we design an intuitive friendly user interface for the potential cloud server with GPUs. Use the user interface through network, we can send the input data to remote server without cumbersome setting in local host. Finally, we can receive the alignment results from the remote cloud server with GPUs.

Recently, many new next-generation sequencing techniques have been proposed. These techniques can produce lot of short reads rapidly. Hence, a number of tools have been developed to map these short reads to the genome. However, with more and more reads sequenced and the length of reads increases, these tools require high memory usage and huge computational cost and are also impractical for utilization. As the GPU has become increasingly more powerful and ubiquitous, many scientific applications have been implemented to enhance the computational performance on GPU platform. In this paper, we proposed a method, CUDA-FRESCO, to map the short reads to the genome by using CUDA on GPU platform. The experimental results present that CUDA-FRESCO can achieve dramatic speed up than other tools. CUDA-FRESCO can be alternative tool for biologists to map the short reads fast.

BACKGROUND: A classical problem in studying genome rearrangements is understanding the series of rearrangement events involved in transforming one genome into another in accordance with the parsimonious principle when two genomes with the same set of genes differ in gene order. The most studied event is the reversal, but an increasing number of reports have considered reversals along with other genome rearrangement events. Some recent studies have investigated the use of reversals and block-interchanges simultaneously with a weight proportion of 1:2. However, there has been less progress towards exploring additional combinations of weights. RESULTS: In this paper, we present several approaches to examine genome rearrangement problems by considering reversals and block-interchanges together using various weight assignments. An exact algorithm for the weight proportion of 1:2 is developed, and then, its idea is extended to design approximation algorithms for other weight assignments. The results of our simulations suggest that the performance of our approximation algorithm is superior to its theoretical expectation. CONCLUSION: If the weight of reversals is no more than that of block-interchanges, our algorithm provides an acceptable solution for the transformation of two permutations. Nevertheless whether there are more tractable results for studying the two events remains open.

Sequence alignment is a fundamental problem in the computational biology. Many alignment methods have been proposed in the literature, such as pair-wise sequence alignment (2SA), syntenic alignment, multiple sequence alignment (MSA) and constraint multiple sequence alignment, etc. Three-sequence alignment (3SA) problem has been proposed and discussed in the computational biology and proved that the alignment results from 3SA are better than those from 2SA under some conditions. However, 3SA problem is less discussed over the past decade due to the computer capability. 3SA problem now is worthy to discuss due to the powerful computer and more and more genome and protein sequences. In this paper, an efficient parallel algorithm (P3SA) is proposed to solve 3SA problem. The P3SA method requires 0(n2/p) space complexity and 0(n3/p) time complexity. The experimental results show that P3SA algorithm is applicable and achieves a satisfied speed-up.

The classification of protein structures is essential for their function determination in bioinformatics. The success of the protein structure classification depends on two factors: the computational methods used and the features selected. In this paper, we use a combinatorial fusion analysis technique to facilitate feature selection and combination for improving predictive accuracy in protein structure classification. When applying these criteria to our previous work, the resulting classification has an overall prediction accuracy rate of 87% for four classes and 69.6% for 27 folding categories. These rates are significantly higher than our previous work and demonstrate that combinatorial fusion is a valuable method for protein structure classification.

In order to support Internet security, virtual private networks, QoS, etc., Internet routers need to classify incoming packets quickly into flows. A packet classifier uses information contained in the packet header and a predefined rule table in the routers to classify the packets. This paper presents a novel packet classification algorithm, called the bit compression algorithm. Like the previously best known algorithm, bitmap intersection, bit compression is based on the multiple dimensional range lookup approach. Since the bit vectors of the bitmap intersection contain lots of '0' bits, the bit vectors could be compressed. We compress the bit vectors by preserving useful information but removing the redundant '0' bits of the bit vectors. Additionally, the wildcard rules also enable more extensive improvement. Comparing with the bitmap intersection algorithm, the bit compression algorithm reduces the storage complexity in the average-case from thetas (dN2) to thetas (dN-logN), where d denotes the number of dimensions and N represents the number of rules. By exploring the memory hierarchy, we show that bit compression algorithm requires much less memory access than bitmap intersection algorithm on Intel IXP1200 network processor. Since memory access dominates the lookup time, even though extra decompression time is required for bit compression scheme, the bit compression scheme in the average still outperforms bitmap intersection scheme on the classification performance

We have proposed the extended Karnaugh map representation (EKMH) scheme for multidimensional array representation. We propose two data compression schemes, EKMR compressed row/column storage (ECRS/ECCS), for multidimensional sparse arrays based on the EKMR scheme. To evaluate the proposed schemes, we compare them to the CRS/CCS schemes. Both theoretical analysis and experimental tests were conducted. In the theoretical analysis, we analyze the CRS/CCS and the ECRS/ECCS schemes in terms of the time complexity, the space complexity, and the range of their usability for practical applications. In experimental tests, we compare the compressing time of sparse arrays and the execution time of matrix-matrix addition and matrix-matrix multiplication based on the CRS/CCS and the ECRS/ECCS schemes. The theoretical analysis and experimental results show that the ECRS/ECCS schemes are superior to the CRS/CCS schemes for all the evaluated criteria, except the space complexity in some case.

Array operations are useful in a large number of important scientific codes, such as molecular dynamics, finite element methods, climate modeling, atmosphere and ocean sciences, etc. In our previous work, we have proposed a scheme of extended Karnaugh map representation (EKMR) for multidimensional array representation. We have shown that sequential multidimensional array operation algorithms based on the EKMR scheme have better performance than those based on the traditional matrix representation (TMR) scheme. Since parallel multidimensional array operations have been an extensively investigated problem, we present efficient data parallel algorithms for multidimensional array operations based on the EKMR scheme for distributed memory multicomputers. In a data parallel programming paradigm, in general, we distribute array elements to processors based on various distribution schemes, do local computation in each processor, and collect computation results from each processor. Based on the row, column, and 2D mesh distribution schemes, we design data parallel algorithms for matrix-matrix addition and matrix-matrix multiplication array operations in both TMR and EKMR schemes for multidimensional arrays. We also design data parallel algorithms for six Fortran 90 array intrinsic functions: All, Maxval, Merge, Pack, Sum, and Cshift. We compare the time of the data distribution, the local computation, and the result collection phases of these array operations based on the TMR and the EKMR schemes. The experimental results show that algorithms based on the EKMR scheme outperform those based on the TMR scheme for all test cases.

For sparse array operations, in general, the sparse arrays are compressed by some data compression schemes in order to obtain better performance. The Compressed Row/Column Storage (CRS/CCS) schemes are the two common used data compression schemes for sparse arrays in the traditional matrix representation (TMR). When extended to higher dimensional sparse arrays, array operations using the CRS/CCS schemes usually do not perform well. We propose two data compression schemes, extended Karnaugh map representation Compressed Row/Column Storage (ECRS/ ECCS) for multi-dimensional sparse arrays based on the EKMR scheme. To evaluate the proposed schemes, both theoretical analysis and experimental tests are conducted. In theoretical analysis, we analyze CRS/CCS and ECRS/ECCS schemes in terms of the time complexity, the space complexity, and the range of their usability for practical applications. In experimental test, we compare the performance of matrix-matrix addition and matrix-matrix multiplication sparse array operations that use the CRS/CCS and ECRS/ECCS schemes. The experimental results show that sparse array operations based on the ECRS/ECCS schemes outperform those based on the CRS/CCS schemes for all test samples.

Array operations are used in a large number of important scientific codes. To implement these array operations efficiently, many methods have been proposed in the literature, most of which are focused on two-dimensional arrays. When extended to higher dimensional arrays, these methods usually do not perform well. Hence, designing efficient algorithms for multidimensional array operations becomes an important issue. We propose a new scheme, extended Karnaugh map representation (EKMR), for the multidimensional array representation. The main idea of the EKMR scheme is to represent a multidimensional array by a set of two-dimensional arrays. Hence, efficient algorithm design for multidimensional array operations becomes less complicated. To evaluate the proposed scheme, we design efficient algorithms for multidimensional array operations, matrix-matrix addition/subtraction and matrix-matrix multiplications, based on the EKMR and the traditional matrix representation (TMR) schemes. Theoretical and experimental tests for these array operations were conducted. In the experimental test, we compare the performance of intrinsic functions provided by the Fortran 90 compiler with those based on the EKMR scheme. The experimental results show that the algorithms based on the EKMR scheme outperform those based on the TMR scheme and those provided by the Fortran 90 compiler.