•The speed of your CPU determines how
quickly it can process instructions
•Many bioinformatics operations fall
into the “embarrassingly parallel” category
•Getting a results faster is as
simple as adding more CPUs
•=> clusters
•Though clusters still dominate,
increasing attention is being paid to servers with large number of cores CPU
For current throughputs, you will need ~8 <dual quad core> nodes
per sequencer to handle raw
data.
For
data analysis, 1-2 nodes is enough for general purpose.
How about use Graphic Processing Unit (GPU) instead?