TheInternetisamassiveglobalnetworkofover700millionusersanditisadding users at the rate of 300,000 per day. This large, distributed, and everchanging network poses a challenge to researchers: How does one study, model, or und- stand such a decentralized, constantly evolving entity? Research in large-scale networks seeks to address this question, and the unique nature of these networks calls for a range of techniques from a host of disciplines. The workshop Com- natorial and Algorithmic Aspects of Networking and the Internet (CAAN 2004) provided a forum for the exchange of ideas on these topics. The primary goals of the workshop were to bring together a diverse cro- section of researchers in an already scattered and distinct community and also to provide a snapshot of the cutting-edge research in this ?eld. We succeeded in these goals: among the participants were mathematicians, computer scientists in theory and algorithms, computer scientists in networks, physicists, and en- neers, as well as researchers from Europe and North America, participants from industry and academia, students, and established researchers; and among the papers were some new and surprising results as well as some introductions to the foundations of the ?eld. Theworkshopprogramfeatured12peer-reviewedpapersbracketedbytwoho- longinvitedsurveytalks anopeningtalkbyAshishGoelandaclosingtalkby- dreiBroder.TopicscoveredbythetalksrangedfromtheWebgraphtogametheory tostringmatching,allinthecontextoflarge-scalenetworks.Thisvolumecollects togetherthetalksdeliveredattheworkshopalongwithanumberofsurveyarticles toroundoutthepresentationandgiveacomprehensiveintroductiontothetopic.