scholarly journals Some Algorithms for Computing Short-Length Linear Convolution

Electronics ◽  
2020 ◽  
Vol 9 (12) ◽  
pp. 2115
Author(s):  
Aleksandr Cariow ◽  
Janusz P. Paplinski
Keyword(s):  
Data Processing ◽  
Energy Efficient ◽  
Hardware Implementation ◽  
Building Blocks ◽  
Short Length ◽  
Complex Data ◽  
Linear Convolution ◽  
Naive Approach

In this article, we propose a set of efficient algorithmic solutions for computing short linear convolutions focused on hardware implementation in VLSI. We consider convolutions for sequences of length N= 2, 3, 4, 5, 6, 7, and 8. Hardwired units that implement these algorithms can be used as building blocks when designing VLSI -based accelerators for more complex data processing systems. The proposed algorithms are focused on fully parallel hardware implementation, but compared to the naive approach to fully parallel hardware implementation, they require from 25% to about 60% less, depending on the length N and hardware multipliers. Since the multiplier takes up a much larger area on the chip than the adder and consumes more power, the proposed algorithms are resource-efficient and energy-efficient in terms of their hardware implementation.

scholarly journals Algorithmic Structures for Realizing Short-Length Circular Convolutions with Reduced Complexity

Electronics ◽  
2021 ◽  
Vol 10 (22) ◽  
pp. 2800
Author(s):  
Aleksandr Cariow ◽  
Janusz P. Paplinski
Keyword(s):  
Energy Efficient ◽  
Hardware Implementation ◽  
Circulant Matrix ◽  
Short Length ◽  
Convolution Operation ◽  
Naive Method ◽  
The Matrix ◽  
Reduced Complexity ◽  
Matrix Vector ◽  
Naive Approach

A set of efficient algorithmic solutions suitable to the fully parallel hardware implementation of the short-length circular convolution cores is proposed. The advantage of the presented algorithms is that they require significantly fewer multiplications as compared to the naive method of implementing this operation. During the synthesis of the presented algorithms, the matrix notation of the cyclic convolution operation was used, which made it possible to represent this operation using the matrix–vector product. The fact that the matrix multiplicand is a circulant matrix allows its successful factorization, which leads to a decrease in the number of multiplications when calculating such a product. The proposed algorithms are oriented towards a completely parallel hardware implementation, but in comparison with a naive approach to a completely parallel hardware implementation, they require a significantly smaller number of hardwired multipliers. Since the wired multiplier occupies a much larger area on the VLSI and consumes more power than the wired adder, the proposed solutions are resource efficient and energy efficient in terms of their hardware implementation. We considered circular convolutions for sequences of lengths N= 2, 3, 4, 5, 6, 7, 8, and 9.

scholarly journals Low-Power Multiplexer Structures Targeting Efficient QCA Nanotechnology Circuit Designs

Electronics ◽  
2021 ◽  
Vol 10 (16) ◽  
pp. 1885
Author(s):  
Amjad Almatrood ◽  
Aby K. George ◽  
Harpreet Singh
Keyword(s):  
Energy Efficient ◽  
Building Blocks ◽  
Switching Frequency ◽  
Circuit Implementation ◽  
Suitable Candidate ◽  
Quantum Dot Cellular Automata ◽  
Performance Factors ◽  
Integration Density ◽  
Qca Circuits ◽  
Energy Efficient Building

Quantum-dot cellular automata (QCA) technology is considered to be a possible alternative for circuit implementation in terms of energy efficiency, integration density and switching frequency. Multiplexer (MUX) can be considered to be a suitable candidate for designing QCA circuits. In this paper, two different structures of energy-efficient 2×1 MUX designs are proposed. These MUXes outperform the best existing design in terms of power consumption with approximate reductions of 26% and 35%. Moreover, similar or better performance factors such as area and latency are achieved compared to the available designs. These MUX structures can be used as fundamental energy-efficient building blocks for replacing the majority-based structures in QCA. The scalability property of the proposed MUXes is excellent and can be used for energy-efficient complex QCA circuit designs.

scholarly journals Using Hadoop Technology to Overcome Big Data Problems by Choosing Proposed Cost-efficient Scheduler Algorithm for Heterogeneous Hadoop System (BD3)

2020 ◽  
pp. 58-84
Author(s):  
Abou_el_ela Abdou Hussein
Keyword(s):  
Big Data ◽  
Data Processing ◽  
Data Storage ◽  
Database Management System ◽  
Data Sets ◽  
Complex Data ◽  
Daily Data ◽  
Complex Data Sets ◽  
Cost Efficient ◽  
Hadoop System

Day by day advanced web technologies have led to tremendous growth amount of daily data generated volumes. This mountain of huge and spread data sets leads to phenomenon that called big data which is a collection of massive, heterogeneous, unstructured, enormous and complex data sets. Big Data life cycle could be represented as, Collecting (capture), storing, distribute, manipulating, interpreting, analyzing, investigate and visualizing big data. Traditional techniques as Relational Database Management System (RDBMS) couldn’t handle big data because it has its own limitations, so Advancement in computing architecture is required to handle both the data storage requisites and the weighty processing needed to analyze huge volumes and variety of data economically. There are many technologies manipulating a big data, one of them is hadoop. Hadoop could be understand as an open source spread data processing that is one of the prominent and well known solutions to overcome handling big data problem. Apache Hadoop was based on Google File System and Map Reduce programming paradigm. Through this paper we dived to search for all big data characteristics starting from first three V's that have been extended during time through researches to be more than fifty six V's and making comparisons between researchers to reach to best representation and the precise clarification of all big data V’s characteristics. We highlight the challenges that face big data processing and how to overcome these challenges using Hadoop and its use in processing big data sets as a solution for resolving various problems in a distributed cloud based environment. This paper mainly focuses on different components of hadoop like Hive, Pig, and Hbase, etc. Also we institutes absolute description of Hadoop Pros and cons and improvements to face hadoop problems by choosing proposed Cost-efficient Scheduler Algorithm for heterogeneous Hadoop system.

Energy-Efficient Data Processing at Sweet Spot Frequencies

2014 ◽  
pp. 154-171 ◽  
Author(s):  
Sebastian Götz ◽  
Thomas Ilsche ◽  
Jorge Cardoso ◽  
Josef Spillner ◽  
Uwe ASSmann ◽  
...  
Keyword(s):  
Data Processing ◽  
Energy Efficient ◽  
Sweet Spot ◽  
Efficient Data

ACTOR: Active Cloud Storage with Energy-Efficient On-Drive Data Processing

2018 ◽  
Author(s):  
Zhi Qiao ◽  
Shuwen Liang ◽  
Nandini Damera ◽  
Song Fu ◽  
Hsing-bung Chen ◽  
...  
Keyword(s):  
Data Processing ◽  
Energy Efficient ◽  
Cloud Storage

scholarly journals In Situ Mass Spectrometry Diagnostics of Impaired Glucose Tolerance Using Label-Free Metabolomic Signature

Diagnostics ◽  
2020 ◽  
Vol 10 (12) ◽  
pp. 1052
Author(s):  
Petr G. Lokhov ◽  
Oxana P. Trifonova ◽  
Dmitry L. Maslov ◽  
Elena E. Balashova
Keyword(s):  
Mass Spectrometry ◽  
Glucose Tolerance ◽  
Data Processing ◽  
Impaired Glucose Tolerance ◽  
Characteristic Curve ◽  
Complex Data ◽  
Label Free ◽  
Mass Peak ◽  
Mass Spectrometers

In metabolomics, mass spectrometry is used to detect a large number of low-molecular substances in a single analysis. Such a capacity could have direct application in disease diagnostics. However, it is challenging because of the analysis complexity, and the search for a way to simplify it while maintaining the diagnostic capability is an urgent task. It has been proposed to use the metabolomic signature without complex data processing (mass peak detection, alignment, normalization, and identification of substances, as well as any complex statistical analysis) to make the analysis more simple and rapid. Methods: A label-free approach was implemented in the metabolomic signature, which makes the measurement of the actual or conditional concentrations unnecessary, uses only mass peak relations, and minimizes mass spectra processing. The approach was tested on the diagnosis of impaired glucose tolerance (IGT). Results: The label-free metabolic signature demonstrated a diagnostic accuracy for IGT equal to 88% (specificity 85%, sensitivity 90%, and area under receiver operating characteristic curve (AUC) of 0.91), which is considered to be a good quality for diagnostics. Conclusions: It is possible to compile label-free signatures for diseases that allow for diagnosing the disease in situ, i.e., right at the mass spectrometer without complex data processing. This achievement makes all mass spectrometers potentially versatile diagnostic devices and accelerates the introduction of metabolomics into medicine.

Sign in / Sign up

Export Citation Format

Share Document