Ling Liu's SC13 paper "Large Graph Processing Without the Overhead" featured by HPCwire.
Another list highlighting Open Source Software Releases.
Second GraphLab workshop should be even bigger than the first! GraphLab is a new programming framework for graph-style data analytics.
Distance-aware bloom filters: Enabling collaborative search for efficient resource discovery
USENIX ;login:, 2013.
Yiming Zhang, Ling Liu*
National University of Defense
Technology (NUDT), Changsha, Hunan, China
*Georgia Institute of Technology
Resource discovery in large-scale Peer-to-Peer (P2P) networks is challenging due to lack of effective methods for guiding queries. Based on the observation that the effectiveness of P2P resource discovery is determined by the utilization of hints, i.e., a summary of where the resources are, scattered in the network, in this paper we propose the distance-aware bloom filters (DABF) that disseminate hint information to faraway nodes by decaying BFs with different deterministic masks. Based on DABF, we design a novel Collaborative P2P Search (CPS) mechanism, which supports intelligent message behaviours including suspend, resume, terminate, move, reside, dispatch, notify and order. The effectiveness of our proposals is demonstrated through theoretical analysis and extensive simulations, in which we observed a remarkable reduction in search latency over previous approaches.
KEYWORDS: Peer-to-Peer; distance-aware bloom filters (DABF); collaborative search; information aggregation; resource discovery
FULL PAPER: pdf