Benchmarking the Elliptic Curve Digital Signature Algorithm and RSA in Key Signing and Verification Operations with Parallelism

• Published in: 2023 Congress in Computer Science, Computer Engineering, & Applied Computing (CSCE) pp. 2521 - 2527
• Main Authors: Dimitoglou, George, Jim, Carol
• Format: Conference Proceeding
• Language: English
• Published: IEEE 24.07.2023
• Subjects: Benchmark testing; Cryptography; Degradation; Elliptic curve cryptography; Elliptic curves; Encryption; Parallel processing; Resistance; RSA
• DOI: 10.1109/CSCE60160.2023.00405

Cryptographic techniques play a crucial role in securing information, and selecting an appropriate cryptographic technique involves considering factors such as encryption strength and computational overhead. Asymmetric cryptographic protocols and algorithms, such as RSA and those based on elliptic curve cryptography, are widely used for their flexibility, encryption strength, and key management. This study focuses on comparing the performance of the Elliptic Curve Digital Signature Algorithm (ECDSA) and RSA, specifically analyzing the impact of parallelism on signing and verification operations. The experimental results demonstrate that ECDSA is faster in signing but slower in verification compared to RSA. However, RSA's performance declines rapidly with increasing key sizes, while ECDSA exhibits a more graceful degradation. The study highlights the advantages of using elliptic curves in terms of reduced key sizes and the potential for ECDSA to become a preferred solution as security requirements increase. The paper's uniqueness is based on the impact of parallel processing on the performance of both cryptosystems. Future work involves further investigating the observed performance patterns and examining the implications of post-quantum resistance.