A　bottom-up　computing　paradigm,　namely　Autonomy-Oriented　Computing　(AOC),　offers　some　principles　and　mechanisms　for　non-linearly　solving　computationally　hard　problems.　The　goal　of　this　paper　is　to　describe　how　AOC-based　emergent　computing　can　be　realized　in　large-scale,　dynamic,　distributed　and/or　decentralized　networks.　A　case　study　is　given　to　demonstrate　the　main　process　and　contribution　of　self-organization,　the　core　of　AOC,　in　a　distributed　local　search　algorithm.　©　2009　IEEE.