Loading the document
1 free read left. Create a free account for unlimited reading + AI research.
Create free accountWhitepaperElectronics
This paper, by Kwangsik Shin and colleagues from ETRI and Inha University, proposes an online gaming traffic generator that reproduces gamer behavior. The authors analyzed packet size and inter departure time (IDT) distributions for two games: Left4Dead (L4D), a cooperative first-person shooter, and World of Warcraft (WOW), a massively multiplayer online role-playing game. The generator produces IDTs and packets from an analytical model of gamer behavior and transmits each packet after its IDT expires. The paper was published in ICEC 2010.
The analysis found clear differences between genres. Most L4D IDTs are under 50 ms, while WOW has many packets with long IDTs, and WOW packets are smaller than L4D packets. In initial tests, generated L4D traffic closely matched the analytical model, but WOW traffic diverged substantially. The authors attribute this to TCP's Nagle algorithm and delayed acknowledgments. After disabling Nagle's algorithm via TCP_NODELAY, the generated WOW distributions moved close to the model, according to the authors, who report that Q-Q plots indicate goodness of fit. They note that WOW IDTs remain less accurate than L4D's because TCP's other control mechanisms still distort timing.
Each row is a game or company this document names, each column a document, oldest first. A mark means that document names it. Select a row to see those documents.
DiGRA · 2022