Performance analysis of data compression algorithms for heterogeneous architecture through parallel approach

Today, there is a huge demand for data compression due to the need to reduce the transmission time and increase the capacity of data storage. Data compression is a technique which represents an information, images, video files in a compressed or in a compact format. There are various data compressio...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of supercomputing Vol. 76; no. 4; pp. 2275 - 2288
Main Authors: Mahammad, Farooq Sunar, Viswanatham, V. Madhu
Format: Journal Article
Language:English
Published: New York Springer US 01.04.2020
Springer Nature B.V
Subjects:
Algorithms
Application programming interface
Arithmetic
Bandwidths
Compilers
Compression ratio
Compression tests
Computer architecture
Computer Science
Data compression
Data storage
Image compression
Interpreters
Lossless equipment
Performance evaluation
Processor Architectures
Programming Languages
Random access memory
Time compression
Video compression
K-RLE algorithm
OpenMP
Data compression
Parallel processing
Arithmetic coding
LZ77 algorithm
Multicore architecture
ISSN:0920-8542, 1573-0484
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Today, there is a huge demand for data compression due to the need to reduce the transmission time and increase the capacity of data storage. Data compression is a technique which represents an information, images, video files in a compressed or in a compact format. There are various data compression techniques which keep information as accurately as possible with the fewest number of bits and send it through communication channel. Arithmetic algorithm, Lempel–Ziv 77 (LZ77) and run length encoding with a K-precision (K-RLE) algorithms are lossless data compression algorithms which have lower performance rate because of their processing complexity as well as execution time. This paper presents an efficient parallel approach to reduce execution time for compression algorithms. The proposed OpenMP is an efficient tool for programming within parallel shared-memory environments. Finally, it shows that performance parallel model experimented using Open Multi-Processing (OpenMP) Application Programming Interface through Intel Parallel studio on multicore architecture platform with spec of Core 2 duo—2.4 GHz, 1 Gb RAM machine of parallel approach for compression algorithms has been improved remarkably against sequential approach. The improvement in compression ratio through an efficient parallel approach leads to reduction on transmission cost, reduction in storage space and bandwidth without additional hardware infrastructure. An overall performance evaluation shows arithmetic data compression algorithm with 46% which is better than LZ77 of 44% as well as K-RLE of 37% data compression algorithms.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0920-8542
1573-0484
DOI:10.1007/s11227-018-2478-3