Year: 2025
Type: preprint
Abstract: This paper addresses critical challenges in bug detection for contemporary cloud-native applications, particularly those developed in Go. We introduce an innovative greybox fuzzing framework that integrates a fine-grained coverage metric with an adaptive scheduling and intelligent seed selection mechanism. This approach effectively mitigates common issues such as seed explosion, which are prevalent in large-scale cloud software. Through extensive empirical validation on live cloud infrastructures, our method demonstrates substantial improvements in bug discovery rates and code coverage compared to conventional fuzzing techniques, thereby significantly bolstering the security assurance of cloud-deployed services. Adaptive Fuzzing
Author: Venu Kalluru
Language: English
Subfield: Software
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4413919737