InfoVis Infrastructure Workshop

1
InfoVis Infrastructure Workshop

• Chris Mueller
• Open Systems Lab, Indiana University
• October 9, 2004
• chemuell at cs dot indiana dot edu
• www.osl.iu.edu

2
Overview

• Position Paper
• Repository style infrastructure (SourceForge,
  GenBank, CPAN)
• Standard software protocols
• Guiding policies to help ensure quality
• Current Work Open Systems Lab, IU
• High performance components for IVC
• Boost Graph Library
• Very large data sets/visualization
• Interests
• Understand community needs
• Learn whats available, where were going
• Industry Viewpoint
• Web-based Scientific Visualization and Analysis
  products
• In-house visualization and analysis tools
  (high-throughput analytical chemistry)

3
Boost Graph Library Algorithms
Core Algorithm Patterns breadth_first_search
breadth_first_visit depth_first_search
depth_first_visit undirected_dfs Shortest
Paths Algorithms dijkstra_shortest_paths
bellman_ford_shortest_paths dag_shortest_paths
johnson_all_pairs_shortest_paths Minimum
Spanning Tree Algorithms kruskal_minimum_spannin
g_tree prim_minimum_spanning_tree
connected_components strong_components
Incremental Connected Components
initialize_incremental_components
incremental_components same_component
component_index
Maximum Flow Algorithms edmunds_karp_max_flow
push_relabel_max_flow topological_sort
transitive_closure copy_graph
transpose_graph isomorphism
cuthill_mckee_ordering sequential_vertex_colori
ng minimum_degree_ordering sloan_ordering
ith_wavefront, max_wavefront, aver_wavefront, and
rms_wavefront
Recent Additions Betweenness Centrality
Betweenness Centrality clustering A search
Floyd-Warshall all-pairs shortest paths
Kamada-Kawai layout
4
Actor Collaboration Database Betweenness
Centrality Clustering (threshold0.01) Single
Processor
5
Actor Collaboration Database Betweenness
Centrality
6
Dot Plot Performance Results

• Base is a direct port of the DOTTER algorithm
• SIMD 1 is the SIMD algorithm using a sparse
  matrix data structure based on STL vectors
• SIMD 2 is the SIMD algorithm using a binary
  format and memory mapped output files
• Thread is the SIMD 2 algorithm on 2 Processors