\@writefile{toc}{\contentsline{subsubsection}{\numberline{\unhbox\voidb@x \hbox{III-C}1}Flag:two clients want to access the same seed simultaneously.}{4}}
\@writefile{toc}{\contentsline{subsubsection}{\numberline{\unhbox\voidb@x \hbox{III-C}2}Service:some clients update bitmap in the database simultaneously}{4}}
\@writefile{lof}{\contentsline{figure}{\numberline{3}{\ignorespaces race of updating bitmap from different clients}}{4}}
This document is a model and instructions for \LaTeX.
This and the IEEEtran.cls file define the components of your paper [title, text, heads, etc.]. *CRITICAL: Do Not Use Symbols, Special Characters, Footnotes,
or Math in Paper Title or Abstract.
Fuzz is an effective technology in software testing and security vulnerability detection. Unfortunately, fuzzing is an extremely compute-intensive job, which may cause thousands of computing hours to find a bug. Current novel works generally improve fuzzing efficiency by developping delicate algorithms. In this paper, we propose another direction of improvement in this filed, i.e., leveraging parallel computing to improve fuzzing effiency. In this way, we develop p-fuzz, a parallel fuzzing system that can utilize massive distributed computing resources to fuzz a single program. p-fuzz uses a no-sql database to share the fuzzing status such as seeds, covered paths, etc. All fuzzing nodes get jobs from the database, and update their fuzzing status to the database. We control the synchronization period to a coarse granularity so that the database will not be a bottleneck. P-fuzz is implemented based on AFL. We compare p-fuzz with AFL and XXX in our experiment. The result shows that we can easily gain a speedup of AFL by simply using 4 nodes, i.e., using 3X more resources.
