Closed karel-brinda closed 6 days ago
These are my values from the human genome, k=31, index construction:
experiment run real_s real_s_py user_s sys_s percent_cpu max_ram_kb fs_inputs fs_outputs exit_code command
None None 975.17 975.179138 955.58 15.56 99% 46845014 0 1396768 0 fmsi index -p masked_superstring.mones.fa
There should be two clearly separated pieces of info:
Memory for index running (this is already sh there)
Memory needed for index construction (e.g., for HG) – just a rough idea to know eg what machine to choose on a cluster - high mem vs standard, etc.