Apache Giraph is a iterative graph processing system built for high scalability. For example, it's currently used at Facebook to analyze the social graph formed by users and their. Giraph originated as the Open-source counterpart to Pregel, the graph processing architecture developed at Google and DESC Ribed in a paper. Both systems are inspired by the Bulk synchronous model of Parallel distributed computation by introduced Leslie. Giraph adds several features beyond the basic Pregel model, including master computation, sharded aggregators, edge-orient Ed input, Out-of-core computation, and more. With a steady development cycle and a growing community of users worldwide, giraph are a natural choice for unleashing the Potential of structured datasets at a massive scale.
Principle: Giraph based on Hadoop, will be mapreduce in mapper encapsulation, not using reducer. Multiple iterations are performed in mapper, each iteration equivalent to the superstep in the BSP model. A Hadoop job is equivalent to a BSP job. The underlying structure is shown in the following illustration.
Infrastructure for Giraph
The functions of each section are as follows:
(1) Zookeeper:responsible for computation State
–global State: #superstep
–checkpoint paths, aggregator values, statistics
(2) Master:responsible for coordination
–assigns partitions to workers
–aggregates Aggregator Values
–collects Health Statuses
(3) Worker:responsible for vertices
–invokes active vertices compute () function
–sends, receives and assigns messages
–computes Local aggregation values