Performance of the Parallel Algorithms

Log-log plot of predicted total execution times (Pltotal and P2totai) and communication times (Plcomm and P2comm) on a Linux cluster for the uracil dimer using the cc-pVDZ basis set. Ideal scaling corresponds to a straight line of negative unit slope. 

The difference in speedup curves for these two cases demonstrates that network performance characteristics can have significant impact on the parallel performance of an algorithm. 

Measured efficiencies for the PI and P2 algorithms on a Linux duster for the uradl dimer using the cc-pVDZ basis set. 

---