A graph service for global web entities traversal and reputation evaluation based on HBase
|
講者：Chris Huang, SPN Hadoop Architect, Trend Micro 時段：10:50~11:30 地點：4F國際會議廳 講題：A graph service for global web entities traversal and reputation evaluation based on HBase |
 |
摘要：
In real world threat environment, we collect lots of threat knowledge data and each contains many different threat (web) entities, e.g. domains, IPs, and file SHA1. Most of the cases, threat entities will be observed with relations between them, which also mean malicious behaviors or interaction chains. So we build up a graph model to store all the known threat entities and their relationships on HBase, to provide a service to allow clients to query threat relationships via any given threat entity. We also implemented a graph model traversal API for client code to use conveniently, and graph algorithms to calculate malicious score of any given threat entity derived from scores of unknown and related threat entities for other reputation systems to references. In this sharing, we'd like to talk about, 1)what problems we try to solve, 2)what and how the design decisions we made, 3)how we design this graph model, and 4)graph computation tasks. In the end, 5)we also talk about our current status and 6) future enhancements.
講者簡介：
Chen-hsiu (Chris) Huang is engaged in constructing Big Data infrastructure with Hadoop ecosystems, and building analytic platform on top of it. The major objective is to enhance enterprise solution efficiency by connecting existing IT systems with business values which are generated from enterprise data analytics.