next up previous
Next: Parallel Iterative Solver Up: Network Ordering Performance Previous: Network Ordering Performance

Parallel Direct Solver Statistics

Statistics are presented in tables gif through gif for four orderings of each of the five matrices used through out this research. In this table, is the number of rows/columns in the borders and last diagonal block of the ordered matrix and is the number of fillin. The matrix partitionings that yielded the best empirical performance during benchmarking the parallel software implementations on the Thinking Machines CM-5 are labeled.

 
Table: BCSPWR09 --- LU and Choleski Factorization Ordering Statistics  

 
Table: BCSPWR10 --- LU and Choleski Factorization Ordering Statistics  

 
Table: EPRI6K --- LU and Choleski Factorization Ordering Statistics  

 
Table: NiMo-OPS --- LU and Choleski Factorization Ordering Statistics  

 
Table: NiMo-PLANS --- LU and Choleski Factorization Ordering Statistics  



David P. Koester
Sun Oct 22 17:27:14 EDT 1995