This sparse matrix visualization was generated by Emily, a
software tool developed by former Indiana University doctoral student Tom
Loos and Randall Bramley. The white lines indicate how the matrix was
partitioned among 64 parallel processors. The colors represent the
magnitudes of the coefficients in the system of linear equations being
solved. Red dots indicate coefficients that were 18 orders of magnitude
greater than those indicated by purple dots. The low ratio of purple to
red dots was a visual clue that can help users find a solution strategy
appropriate to this particular matrix -- something that would have been more
difficult to discover based on numbers alone. Bramley and his colleagues
plan to incorporate Emily as a component of the Linear System Analyzer.
This sparse matrix visualization was generated by Emily, a
software tool developed by former Indiana University doctoral student Tom
Loos and Randall Bramley. The white lines indicate how the matrix was
partitioned among 64 parallel processors. The colors represent the
magnitudes of the coefficients in the system of linear equations being
solved. Red dots indicate coefficients that were 18 orders of magnitude
greater than those indicated by purple dots. The low ratio of purple to
red dots was a visual clue that can help users find a solution strategy
appropriate to this particular matrix -- something that would have been more
difficult to discover based on numbers alone. Bramley and his colleagues
plan to incorporate Emily as a component of the Linear System Analyzer.
