This project aims to enhance the capabilities of the Host Identity Protocol version 2 (HIPv2) within the Mininet network emulator and PyHIP framework. The objectives include integrating HIPv2 into Mininet to facilitate testing of HIPv2 environments and optimizing PyHIP for real-world applications. Integration with Mininet involves running the HIP daemon within a virtual host environment and ensuring successful ping replies between HIP-enabled Mininet hosts. Optimization efforts focus on potentially changing cryptographic algorithms. These enhancements aim to improve the testing and practical usability of HIPv2 in network emulation. The tests conducted for this thesis involved only two hosts, which limits the comprehensiveness regarding scalability and performance in larger network setups. While the initial findings are promising, a more extensive evaluation with varied scenarios and additional hosts is critical to fully grasp the system's capabilities. Regarding performance of the unmodified version, tests reveal that TCP window sizes ranging from 128,000 to 225,000 bytes and buffer settings at 3,000 $bytes$ were optimal for optimizing throughput. The last objective of throughput optimization, transitioning from PyCryptoDome to the cryptography library boosts throughput by approximately 20 Mbits/s, reaching up to 59 Mbits/s. Furthermore, it is important to acknowledge that modifications were not extended to the symmetric cryptography libraries, like AES, which could potentially greatly influence system throughput. Therefore, we recommend to further extend these modifications to validate throughput improvements across all modified cryptography algorithms. Depending solely on a single high-performing setup may mislead the assessment of the changes, as various iPerf3 test combinations could yield different outcomes. Thus, it is essential to recognize that these limitations may impact the credibility of the findings.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:liu-205131 |
Date | January 2024 |
Creators | Christensson, Martin, De Paz, Edwin |
Publisher | Linköpings universitet, Institutionen för datavetenskap |
Source Sets | DiVA Archive at Upsalla University |
Language | English |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0015 seconds