Journal Publication of International Research for Engineering and Management

(JOIREM)
Volume: 10 Issue: 08 | August-2024

Evaluating RSA Key Length: Impact on Security Hardness and Computational Efficiency

Divya1, Upasna Setia2

1
Computer Science and Engineering, Ganga Institute of Technology and Management
---------------------------------------------------------------------***--------------------------------------------------------------
Abstract - The research investigate the impact of advancement by RSA is that it introduced the

key length of RSA on security and computational concept of two keys (public key cryptography), it

efficiency, using simulation we evaluate how key uses two keys: Public Key and Private Key. Public key

length ranging from 1024 to 4096 bits perform for is for encryption and private key is for decryption.

encryption/decryption process and their resistance to

The idea behind the RSA is that it is impossible to
factorization attacks. The findings reveal that longer
factorize large integers. The security of RSA depends
key length enhances the security by increasing the
on its computational complexity. RSA is considered
complexity of factorization. But they also incur
as the foundational element of modern
higher computational cost, also they can affect the
cryptographic system, popularly used for securing
time and resource utilization. The result suggest that
digital communication. With the large key lengths a
2048-bit keys offer good security to general
problem becomes more complex. We need to
application whereas 3072-bit or 4096 bit keys are
understand the relationship between RSA key
recommend for high security environment . We
length, security, and efficiency for maintaining
discuss these findings in the context of emerging
security standards because of computational power
quantum computing threats and propose guidelines
advances and cryptographic or quantum attacks.
in key length selection for future cryptographic
system. The public key consists of two numbers where one

Key Words: Cryptography, Asymmetric algorithm, number is multiplication of two large prime number

RSA ( Rivest, Shamir, and Adleman), Key lengths, and private key is derived from the same two prime

Security, Efficiency. number.RSA key can be typically 1024 or 2048 bits

long .Experts believe that 1024 bits keys can be
1. INTRODUCTION broken in near future, but till now its seems to be
RSA, one of the oldest cryptographic algorithm infeasible task.
designed by Ron Rivest, Adi Shamir and Leonard
Adleman, who publicly described the algorithm in
1977. It is patent in USA on 1983. The significant

© 2024, JOIREM |www.joirem.com| Page 1

 Journal Publication of International Research for Engineering and Management
(JOIREM)
Volume: 10 Issue: 08 | August-2024

 Memory Usage: The amount of

RSA
Plain text
Encryption
memory consumed during
Plain Text
RSA
Decryption encryption and decryption.

The data was logged in structured formats for

Figure 1: RSA Encryption/Decryption
subsequent analysis, ensuring that all relevant
Process
metrics were accurately captured. The collected
2. METHODOLOGY data was analyzed using Python's Pandas library,
The research methodology used to examine the with visualizations created using Matplotlib and
impact of RSA key length on its security and Seaborn. The analysis focused on the following
management. For analyzing the RSA algorithm’s aspects:
security implication and performance an
experimental approach is used. This involves 1. Performance Comparison:

generating RSA keys of various lengths then  Comparing the encryption and decryption

performing encryption and decryptio

decryption operation times across different key lengths
l to

by using these key lengths, measuring the identify trends and trade-offs
trade between

associated computational complexity and security and efficiency.

security strengths. The key length

lengths examined in 2. Resource Utilization Analysis:
Analysis

this study are 1024, 2048, 3072, and 4096 bits.  Analyzing the CPU and memory usage to

Data was collected in two primary categories: understand the computational cost

performance metrics and resource utilization. associated with each key length.
3. Security Implications:
1. Performance Metrics:  Reviewing the theoretical
theoretic and practical
 Encryption Time: The time taken to security
ty provided by each
e key length,
encrypt data with each key length. supported by existing literature and the
 Decryption Time: The time taken performance data collected.
to decrypt data with each key
length. The results of this analysis were used to draw

2. Resource Utilization: conclusions about the optimal RSA key lengths for

 CPU Usage: The percentage of CPU various applications, balancing the need for security

resources consumed
umed during against the constraints of computational resources.

encryption and decryption.

© 2024, JOIREM |www.joirem.com

www.joirem.com| Page 2
 Journal Publication of International Research for Engineering and Management
(JOIREM)
Volume: 10 Issue: 08 | August-2024

Simulation Setup: the average CPU and memory usage for

encryption and decryption operations.
The simulation aims to evaluate the
performance and security of RSA using different By combining C++ and Python we measured and
key lengths(for example 1024, 2048, 3072, and analyzed the performance and resource
4096 bits). We make use of both C++ and utilization with different RSA key lengths,
Python programming languages to implement providing a comprehensive understanding of the
the simulations. trade-offs between security and efficiency of
RSA key lengths.
The simulation setup involves:

Performance Metrics:
 Generation of RSA keys of different length:
Use C++ and the OpenSSL library to
1. Encryption/Decryption Time: We measured
generate RSA keys of different
the time use in encryption and decryption
lengths.
process for each key length.
 Measuring the encryption and decryption
2. Resource Utilization: We evaluated the
time. With the generated RSA keys we
computational resources (CPU, memory)
measure the encryption and decryption time.
consumed during encryption and decryption
With the help of C++ program, encrypt a
processes.
random plaintext, decrypts the cipher text,
3. Security Assessment: We analyzed the
and records the time taken for these
resilience of different key lengths against
operations. The program takes a key file,
factorization attacks algorithms, such as the
encrypts the randomly generated plaintext,
General Number Field Sieve (GNFS).
and the decrypts it, measuring time for both
operations. 3. RESULTS
 Monitoring system resource usage during Encryption/Decryption Performance: Results
these operations. shows increase in encryption and decryption time
To monitor the memory and CPU usage during as key length increases.
encryption and decryption we look to Python
with the ‘psutil’ library. We run python script 1024-bit Keys: Fastest performance but its
with C++ performance measurement program security is insufficient for modern applications.
for each key file as a sub process and calculate

© 2024, JOIREM |www.joirem.com| Page 3

 Journal Publication of International Research for Engineering and Management
(JOIREM)
Volume: 10 Issue: 08 | August-2024

 2048-bit Keys: Acceptable performance with

tough security for general use.
 3072-bit and 4096-bit Keys: Noticeable
increase in processing time, suitable for
high-security environments where
performance trade-offs are acceptable.

Security Analysis: The security analysis

shows that 1024-bit key lengths are
vulnerable to modern factorization Figure 2: Encryption and Decryption Time vs. Key
techniques. 2048 -bit keys provides strong Length.
protection against current computational
complexities. Keys 3072 -bit and 4096-bit 150
CPU
offers greater security, and they are essential 100 Usage(%)
for highly sensitive data. So the analysis 50
concludes that longer key lengths enhance 0 Memory
Usage(MB
the security against the factorization attacks. )

Resource Utilization: Longer key lengths

shows higher CPU and memory usage. The
Figure 3: CPU and Memory Usage vs. Key
difference was noticeable when moving from
Length.
2048-bit to 3072-bit and 4096-bit keys,
highlighting the need for adequate hardware
The security analysis shows that longer key
resources in high-security applications.
significantly enhance the difficulty of factorization
attacks. While 1024-bit keys are sensitive to modern
factorization techniques, 2048 –bit keys give strong
protection against computational capabilities. Keys
greater than these offer even great security, for
highly sensitive data.

Table 1: Encryption/Decryption Performance Discussion Trade –off Between Security and

Metrics for RSA Key lengths Efficiency: The study shows the trade off between

© 2024, JOIREM |www.joirem.com| Page 4

 Journal Publication of International Research for Engineering and Management
(JOIREM)
Volume: 10 Issue: 08 | August-2024

security and computational efficiency in RSA Our research underscores the critical impact of
encryption. While longer key provides better RSA key length on both security and
security, they also demand more computational computational efficiency. While 2048-bit keys are
resources and time foe encryption and decryption suitable for general use, 3072-bit and 4096-bit
operations. keys are necessary for high-security applications.
The findings also highlight the need for ongoing
Current Best Practices: Based on our findings, for evaluation of cryptographic standards in response
most applications 2048 bit keys provides optimal to technological advancements, particularly the
balance, offers good security without any emergence of quantum computing. Future work
performance penalties. 3072 or 4096 bit keys are should focus on developing and adopting
recommended for application which require quantum-resistant cryptographic algorithms to
heightened security, such as financial transaction or ensure long-term security.
government communication despite the increased
computational overhead. REFERENCES
[1] Lenstra, A. K., & Verheul, E. R. (2001).
Implication of Quantum computing: Advancement in
Selecting cryptographic key sizes. Journal of
quantum computing poses a significant threat to RSA
cryptology, 14, 255-293.
encryption, for example Shor’s algorithm can factor
[2] M. Mohan, and J. Prakash, "Analysis of
large integers efficiently. Because of that the
various cryptographic algorithms,"
importance of quantum resistant algorithm will
International Journal of Engineering
increase, longer RSA key lengths provide interim
Technology, Management and Applied
security solutions, until such algorithm become
Sciences, 2(3), 201, pp. 51-61.
widely available.
[3] R. Kumar, and C. C. Ravindranath, "Analysis of
Diffie Hellman Key Exchange Algorithm with
Protection Against Attacks Proper key generation,
proposed Key Exchange Algorithm," Int. J.
including the use of strong random number
Emerg. Trends Technol. Comput. Sci., 4(1), 2015,
generators and secure prime number selection, is
pp. 40-43.
important to ensure that RSA keys are resistant
[4] Mahajan, Prerna, and Abhishek Sachdeva. "A
to attacks such as factorization or mathematical
study of encryption algorithms AES, DES and
vulnerabilities.
RSA for security." Global journal of computer
4. CONCLUSIONS science and technology 13.15 (2013): 15-22.

© 2024, JOIREM |www.joirem.com| Page 5

 Journal Publication of International Research for Engineering and Management
(JOIREM)
Volume: 10 Issue: 08 | August-2024

[5] Bisht, Nivedita, and Sapna Singh. "A Computer Applications(0975-8887) Volume 67-
comparative study of some symmetric and No. 19, April 2013.
asymmetric key cryptography [10] Lim, Meng-Hui, Sanggon Lee, and Sangjae
algorithms." International Journal of Innovative Moon. "Cryptanalysis of Tso et al.’s id-based
Research in Science, Engineering and tripartite authenticated key agreement
Technology 4.3 (2015): 1028-1031. protocol." International Conference on
[6] Arora, Priyanka, Arun Singh, and Himanshu Information Systems Security. Berlin,
Tiyagi. "Evaluation and comparison of security Heidelberg: Springer Berlin Heidelberg, 2007.
issues on cloud computing environment." World [11] Mahajan, Prerna, and Abhishek Sachdeva. "A
of Computer Science and Information study of encryption algorithms AES, DES and
Technology Journal (WCSIT) 2.5 (2012): 179-183. RSA for security." Global journal of computer
[7] Shashi Mehrota Seth, Rajan Mishra, science and technology 13.15 (2013): 15-22.
“Comparative Analysis of Encryption Algorithms [12] Jeeva, A. L., Dr V. Palanisamy, and K.
for Data Communication”, International Journal Kanagaram. "Comparative analysis of
of Computer Science and Technology, Vol. 2, performance efficiency and security measures
Issue 2, pp. 292-294, June 2011. of some encryption algorithms." International
[8] Ajay Kakkar, M.L Singh and P.K. Bansal, Journal of Engineering Research and
“Comparison of Various Encryption Algorithms Applications (IJERA) 2.3 (2012): 3033-3037.
and Techniques for Secured Data [13] Jolly Shah and Dr. Vikas Saxena,” Performance
Communication In Multinode Network”, Study on Image Encryption Schemes” In: IJCSI
International Journal of Engineering and International Journal of Computer Science
Technology Volume 2 No. 1, pp. 87-92, January Issues, Vol. 8, Issue 4.
2012. [14] Gurpreet Singh , Supriya, “A Study of Encryption
[9] Gurpreet Singh , Supriya, “A Study of Encryption Algorithms ( RSA, DES, 3DES and AES) for
Algorithms ( RSA, DES, 3DES and AES) for Information Security”, International Journal of
Information Security”, International Journal of omputer Applications(0975-8887) Volume 67-
No. 19, April 2013.

© 2024, JOIREM |www.joirem.com| Page 6