Linear-time parameterized algorithms with limited local resources

We propose a new computational model for the study of massive data processing. Our measures complexity reading input in terms their very large size N and analyzes cost parameter k that characterizes power provided by limited local computing resources. develop algorithmic techniques solving well-known problems on model. In particular, randomized algorithms running time O(N+g1(k)) space O(k2), with high probability, are developed famous graph matching problem unweighted weighted graphs. More specifically, our algorithm graphs finds k-matching (i.e., edges) general O(N+k2.5), maximum O(N+k3log⁡k).