Parallel Implementation of K-Means Algorithm on FPGA

Dias, Leonardo Alves; Ferreira, Joao Canas; Fernandes, Marcelo A. C.

doi:10.1109/access.2020.2976900

Cited by 16 publications

(18 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the last few years, many researchers have proposed many parallel implementations of the K-means algorithm. A completely parallel implementation of the K-means clustering technique was developed on FPGA [8]. To accelerate processing time and separate huge amounts of data (Big Data) into K clusters, they used the Euclidean distance metric to calculate the similarity between the data and the initial centroid of each cluster to determine to which cluster the data belongs.…”

Section: Related Workmentioning

confidence: 99%

Optimized FPGA-Based Implementation of Brain Tumor Detection by Combining K-Means and Grey Wolf Optimization Algorithms

Jarrah¹,

Amri²

2022

View full text Add to dashboard Cite

There is a need for fast, accurate, and real-time algorithms to detect brain tumors effectively to support the physician's decision-making for treatment purposes. A brain tumor is a lifethreatening uncontrolled growth of cells and tissues that may cause death due to inaccurate and late detection. K-means clustering is one of the clustering techniques that is widely used in brain tumor detection, but it has some drawbacks such as dependency on initial centroid values and a tendency to fall on local optima. This research proposes a new model that uses grey wolf optimization to find the optimal value of K (clusters number) of the k-means algorithm to avoid local optima. A parallel implementation of the K-means clustering algorithm on a field-programmable gate array (FPGA) is also proposed to enhance the performance by reducing the processing time and the power consumption. Moreover, the proposed algorithm is implemented using the Vivado HLS tool on Xilinx Kintex7 XC7K160t FPGA 484-1 where different optimization techniques are adopted and applied, such as loop unrolling, loop pipelining, dataflow, and loop merging. The achieved speed-up of the parallel implementation compared with sequential implementation was 88.17, where the obtained average clustering accuracy was 97.11%.

show abstract

Section: Related Workmentioning

confidence: 99%

Optimized FPGA-Based Implementation of Brain Tumor Detection by Combining K-Means and Grey Wolf Optimization Algorithms

Jarrah¹,

Amri²

2022

View full text Add to dashboard Cite

show abstract

“…Additionally, the amount of data transit is reduced by 28×, relative to conventional kNN. Dias et al also present an implementation for k-means through an HDL based custom design [36]. Design parameters include the bit-width of the fixed-point arithmetic employed, and the number of cores, i.e., replication of the pipelines for distance calculation and cluster assignment for a single data point.…”

Section: Related Workmentioning

confidence: 99%

Optimizing OpenCL Code for Performance on FPGA: k-Means Case Study With Integer Data Sets

2020

View full text Add to dashboard Cite

High Level Synthesis (HLS) tools targeting Field Programmable Gate Arrays (FPGAs) aim to provide a method for programming these devices via high-level abstractions. Initially, HLS support for FPGAs focused on compiling C/C++ to hardware circuits. This raised the issue of determining the programming practices which resulted in the best performing circuits. Recently, to further increase the applicability of HLS approaches, renewed effort was placed on support for HLS of OpenCL code for FPGA, raising the same issues of coding practices and performance portability. This paper explores the performance of OpenCL code compiled for FPGAs for different coding techniques. We evaluate the use of task-kernels versus NDRange kernels, data vectorization, the use of on-chip local memories, and data transfer optimizations by exploiting burst access inference. We present this exploration via a case study of the k-means algorithm, and produce a total of 10 OpenCL implementations of the kernel. To determine the effects of different data set characteristics, and to determine the gains from specialization based on number of attributes, we generated a total of 12 integer data sets. The data sets vary regarding the number of instances, number of attributes (i.e., features), and number of clusters. We also vary the number of processing cores, and present the resulting required resources and operating frequencies. Finally, we execute the same OpenCL code on a 4 GHz Intel i7-6700K CPU, showing that the FPGA achieves speedups up to 1.54× for four cases, and energy savings up to 80 % in all cases.

show abstract

“…FPGAs are widely known for their flexibility for parallelization and low-power consumption. An FPGA is a matrix of logic blocks that allows designing different circuits, such as processors, logic circuits, and even algorithm development [ 31 ]. FPGA platforms can be categorized as third generation computational infrastructure in bioinformatics, as it is a CHA.…”

Section: Introductionmentioning

confidence: 99%

Proposal of Smith-Waterman algorithm on FPGA to accelerate the forward and backtracking steps

2022

Self Cite

View full text Add to dashboard Cite

In bioinformatics, alignment is an essential technique for finding similarities between biological sequences. Usually, the alignment is performed with the Smith-Waterman (SW) algorithm, a well-known sequence alignment technique of high-level precision based on dynamic programming. However, given the massive data volume in biological databases and their continuous exponential increase, high-speed data processing is necessary. Therefore, this work proposes a parallel hardware design for the SW algorithm with a systolic array structure to accelerate the forward and backtracking steps. For this purpose, the architecture calculates and stores the paths in the forward stage for pre-organizing the alignment, which reduces the complexity of the backtracking stage. The backtracking starts from the maximum score position in the matrix and generates the optimal SW sequence alignment path. The architecture was validated on Field-Programmable Gate Array (FPGA), and synthesis analyses have shown that the proposed design reaches up to 79.5 Giga Cell Updates per Second (GCPUS).

show abstract

Parallel Implementation of K-Means Algorithm on FPGA

Cited by 16 publications

References 20 publications

Optimized FPGA-Based Implementation of Brain Tumor Detection by Combining K-Means and Grey Wolf Optimization Algorithms

Optimized FPGA-Based Implementation of Brain Tumor Detection by Combining K-Means and Grey Wolf Optimization Algorithms

Optimizing OpenCL Code for Performance on FPGA: k-Means Case Study With Integer Data Sets

Proposal of Smith-Waterman algorithm on FPGA to accelerate the forward and backtracking steps

Contact Info

Product

Resources

About