I understand that is always better to compile the reference genome indices on a given machine for the different aligners such as BWA, Bowtie2 and HiSat. However, the human genome takes a long time to be processed.
In order to avoid loosing so much computational power, is it possible to re-use (that is, copying them from one machine to another) the indices on machines with the same architecture, let's say AMD processors or Linux 64 bit environment?
What are the restrains one should look at when re-using the indices? otherwise one could simply use the pre-compiled indices available online.