Presentation is loading. Please wait.

Presentation is loading. Please wait.

Fast, Exact Graph Diameter Computation with Vertex Programming Corey Pennycuff and Tim Weninger SIGKDD Workshop on High Performance Graph Mining August.

Similar presentations


Presentation on theme: "Fast, Exact Graph Diameter Computation with Vertex Programming Corey Pennycuff and Tim Weninger SIGKDD Workshop on High Performance Graph Mining August."— Presentation transcript:

1

2 Fast, Exact Graph Diameter Computation with Vertex Programming Corey Pennycuff and Tim Weninger SIGKDD Workshop on High Performance Graph Mining August 10, 2015 Vertex-Centric Computing for Large Scale Graph Analytics

3 Dijkstra’s Single Source Shortest Path A C F E D B 0 2 ABCDEFG A011122 G

4 Medium Graphs 4 million nodes 200 million edges

5 Bigger Graphs Solution – Hadoop data mappers shuffle and sort reducers result 234 DISK

6 Graph Diameter HADIReverse Cuthill-McKeeRandom BFS

7 Bulk Synchronous Parallel (BSP) Created in 1990 by Les Valiant and Bill McColl at Oxford data result Superstep 1 Superstep 2 Superstep 3 Data kept in memory DISK Superstep 0 barrier

8 Graph Analytics with BSP Require the programmer to “think like a vertex” A C F E D B …

9 The Vertex Each Vertex Can: Receive messages from previous superstep Modify its value/datum Send messages

10 BSP Single Source Shortest Path compute(MessageIterator* msgs){ bool changed = false; foreach(msg : msgs){ if(msg < datum){ datum = msg; changed = true; } if(changed) { foreach(edge : GetOutEdgeIterator()){ sendMessageTo(edge.dest, datum + edge.weight) } }else{ voteToHalt(); } A C F E D B G

11 Dijkstra’s Single Source Shortest Path ABCDEFG A0 Superstep 0 master A C F E D B 0 G

12 Dijkstra’s Single Source Shortest Path ABCDEFG A0112 Superstep 1 A C F E D B 0 G

13 Dijkstra’s Single Source Shortest Path Superstep 2 A C F E D B 0 G ABCDEFG A0112

14 Supersteps-1 = Node Eccenctricity A C F E D B 0 G ABCDEFG A0112

15 Diameter Measurement A C F E D B G A C F E D B G A C F E D B G A C F E D B G A C F E D B G A C F E D B G A C F E D B G

16 Limitations Must be synchronous Designed for unweighted graphs

17 Performance Results ER-Graphs (p=32%)

18 Performance Results SF-Graphs (k=3)

19 Performance Results Real World Graphs

20 Thank you


Download ppt "Fast, Exact Graph Diameter Computation with Vertex Programming Corey Pennycuff and Tim Weninger SIGKDD Workshop on High Performance Graph Mining August."

Similar presentations


Ads by Google