DSpace About DSpace Software
 

DSpace on Chitkara University >
Thesis - Chitkara University Himachal Pradesh >
Phd Thesis 2015 HP >

Please use this identifier to cite or link to this item: http://dspace.chitkara.edu.in/xmlui/handle/1/567

Title: PARALLEL ALGORITHMS FOR PUBLIC KEY INFRASTRUCTURE BASED SECURITY TECHNIQUES
Authors: Saxena, Sapna
Keywords: PARALLEL ALGORITHMS
PUBLIC KEY
SECURITY TECHNIQUES
Issue Date: 31-Dec-2014
Publisher: Chitkara University Publications
Abstract: Cryptographic algorithms are used extensively ensuring the security of data when it’s static as well as on the un-trustworthy communication channels. A cryptographic algorithm is the mathematical function used for the encryption and the decryption of data. There are two types of key based data encryption algorithms: symmetric and public Key infrastructure. Symmetric algorithms are those where encryption key can be calculated from decryption key and vice versa. Public key cryptographic algorithms use a pair of two distinct keys – public key and private key - that are used for encryption and decryption or for signing and verification. The main property of public key infrastructure based algorithms is that practically it is impossible to calculate the decryption key from a given encryption key. RSA and RSA-based digital signature algorithms are some of the most popularly used public-key infrastructure cryptographic algorithms with their roots in the modular arithmetic. To strengthen the security of these algorithms, larger key sizes are used typically 1024 bit or more, and that makes these algorithms compute-intensive taking longer time during encryption - decryption and signing - verification modules of secure applications. Since the public key based cryptographic algorithms are compute-intensive, their sequential implementations take a lot of time and energy to execute. Parallelization of these security algorithms in order to distribute the complex computational part among the various cores available with the processors today, will achieve higher performance and also be more energy efficient. The objective of this research is to study and analyze various public key infrastructure based cryptographic algorithms and design new parallel algorithms to implement the public-key algorithms such as the RSA and the Digital Signature Algorithm. This research work has been divided into four main parts – (i) to convert memory-efficient algorithms into new parallel algorithms, (ii) to use important parallel programming techniques to obtain the new high xx iv performance parallel variants of these algorithms, (iii) to use OpenMP API to test the algorithms and analyze performance gained by parallelizing the security algorithms through experiments over large number of data sets and, (iv) to use parallelism to demonstrate more energy-efficient algorithms applicable to portable and mobile devices. During this research work, three parallel algorithms have been developed –PRSA1, PRSA2 and PRSADSA. The PRSA1 and the PRSA2 algorithms are the parallel variants of the RSA algorithm and the PRSADSA is the parallel variant of RSA-based digital signature algorithm. The PRSA1 has shown a speedup of up to 5x approximately, the PRSA2 has shown the speedup of up to 7.5x speedup approximately and the PRSADSA has shown a speedup of up to 5x approximately as compared to their respective sequential counterparts, when executed on 8-core machine. All of the three parallel variants were found to be memory-efficient, time-efficient, scalable and cost-optimal during the experimentation. Moreover, all of the three parallel variants are energy-efficient also. The energy consumption of all parallel variants has been measured using Joulemeter. It was observed during experimentation that they consume less energy as compared to sequential counterparts at the same performance levels. Hence these parallel variants are more suitable to be used on battery-operated mobile devices.
URI: http://dspace.chitkara.edu.in/xmlui/handle/1/567
Appears in Collections:Phd Thesis 2015 HP

Files in This Item:

There are no files associated with this item.

View Statistics

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback