Numerous application areas, including bioinformatics and computational biology, demand increasing amounts of processing capability. In many cases, the computation cores data types are suited to field-programmable gate arrays. The challenge is identifying design techniques that can extract high performance potential from FPGA fabric.

Approximate string matching is fundamental to bioinformatics, and has been the subject of numerous FPGA acceleration studies. We address issues with respect implementations both BLAST- dynamic programming- (DP) based methods. Our primary contributions are two new algorithms for emulating seeding extension phases BLAST. These operate in a single pass through database at streaming rate (110 Maa/sec on VP70 query sizes up 600 170 Virtex4 1024), no preprocessing other than loading string....

GitHub is a widely used collaborative platform for global software development. A pull request plays an important role in bridging code changes with version controlling. Developers can freely and parallelly submit requests to base branches wait the merge of their contributions. However, several developers may edit same lines code; such result latent conflict. We refer that tend change remain open during overlapping time period as competing requests. In this paper, we conduct study on 9,476...

Field-programmable gate arrays are widely considered as accelerators for compute-intensive applications. A critical phase of FPGA application development is finding and mapping to the appropriate computing model. enables models with highly flexible fine-grained parallelism associative operations such broadcast collective response. Several case studies demonstrate effectiveness using these in developing applications molecular modeling.

FPGA-based acceleration of molecular dynamics (MD) has been the subject several recent studies. The paper describes a new non-bonded force computation pipeline implemented on 2004-era COTS FPGA board and its integration into ProtoMol MD code. There are innovations: novel interpolation strategy; introduction "semi-floating point" format; various issues related to system integration. As result, authors able model far more particle types, without relying complex buffering, obtain higher...

FPGA-based acceleration of molecular dynamics (MD) has been the subject several recent studies. Implementing long-range forces, however, only recently addressed. Here we describe a solution based on multigrid method. We show that is, in general, an excellent match to FPGAs: primary operations take advantage large number independently addressable RAMs and efficiency with which complex systolic structures can be implemented. The accelerator integrated into our existing MD system, overall...

Molecular dynamics (MD) is of central importance to computational chemistry. Here we show that MD can be implemented efficiently on a COTS FPGA board, and speed-ups from 31/spl times/ 88/spl over PC implementation obtained. Although the amount speed-up depends stability required, 46/spl obtained with virtually no detriment, upper end range apparently viable in many cases. We sketch our implementations describe effects precision trade-off between performance quality simulation.

We show how an application of central importance to computational biochemistry can be implemented efficiently on FPGA. This requires reformulating the algorithm and taking advantage characteristics maximize device utilization.

Molecular docking is one of the primary computational methods used by pharmaceutical companies to try reduce cost drug discovery. A common technique, for low-resolution screening or as an intermediate step, performs a three-dimensional correlation between two molecules test favorable interactions them. We extend our previous work on FPGA-based accelerators, using reconfigurability customization physical laws and geometric models that describe molecule interaction. Our approach, based direct...

In this paper, we examine retiming and pipelining in the context of system-level optimization techniques. Our main contributions are: (a) functionally equivalent delay balancing as necessary techniques for graphs while maintaining numerical fidelity, (b) clock-rate pipelining, a new technique that leverages knowledge multi-rate design spec to pipeline multi-cycle paths. All these have been implemented within HDL Coder™, tool generates synthesizable code from Simulink ® MATLAB®.

Correlation is a standard technique for recognizing known patterns in two-dimensional grid (pixel) images. Its obvious importance has led to numerous hardware implementations and variations. Images captured directly onto 3D voxel grids are becoming more common, including those from confocal microscopy medical imaging technologies. To our knowledge, no one yet addressed correlation as templates such data. We find that this problem includes number of issues: efficient three-axis rotation...

Crash reproduction, which spends much time of developers in reading and understanding source code, is a crucial yet time-consuming task program debugging. To reduce the resource cost, automatic techniques test generation have been proposed. These aim to automatically generate cases reproduce scenario crashed project. Unfortunately, due lack detailed comprehension generated case may fail reproducing an expected crash. In this paper, we propose approach reproducible bug detection. This...

A highly-configurable system provides many configuration options to diversify application scenarios. The combination of these results in a large search space configurations. This makes the detection configuration-related faults extremely hard. Since it is infeasible exhaust every configuration, several methods are proposed sample subset all configurations detect hidden faults. Configuration sampling can be viewed as process repeating pre-defined action whole space, such one-enabled or...

Credit risk assessment plays an important role in financial institutions and banks. A common method of predicting credit risks is to build interpretable rule-set model distinguish high-risk non-risk users. This further assists experts analysts make critical decisions. In a model, the quality selected rules dramatically influences prediction accuracy. It challenging automatically select for learning task.

Abstract The endpoint control of Basic Oxygen Furnace (BOF) steelmaking depends on the prediction carbon content and temperature. However, predicting these variables is challenging because numerous working conditions in industrial field volatility sensor data collected during BOF steelmaking. accuracy models ensemble learning significantly initial distribution data. complex nature makes it to generate diverse subsets, which ultimately affects predictions. This paper presents a new approach...

Customer segmentation plays a crucial role in credit risk assessment by dividing users into specific levels based on their scores. Previous methods fail to comprehensively consider the stability process, resulting frequent changes and inconsistencies users' over time. This increases potential risks company. To this end, paper at first introduces formalizes concept of regret process. However, evaluating is challenging due its black-box nature computational burden posed vast user data sets....

Molecular dynamics (MD) is of central importance to computational chemistry and its myriad applications. In this paper we show that, at even a preliminary stage development, MD can be implemented efficiently on COTS FPGA board, that 57x speed-up over PC implementation obtained. We sketch our describe how performance tuning precision management could double factor.

Software crashes represent unexpected program interrupts that manifest as software faults and can be dangerous in their frequent occurrence diminishes user experience, damage the reputation of a company, potentially cause significant losses to stakeholders.The increasing scale complexity modern requires methods for preventing/handling crashes.Here, we briefly review summarize progress three areas research field associated with handling crashes: crash analysis, reproduction, localization repair.