uchicago-bio / 2015-Autumn-Forum

0 stars 0 forks source link

Parallelization and access of Python data structures; #60

Closed ghost closed 9 years ago

ghost commented 9 years ago

Okay, so: I have a multiprocessing framework for question #3. I modified the sequence alignment package to use PAM120 and to return a maximum score instead of alignment lists, and I have an algorithm that chunks up the database into memory.

But: The program is designed such that the workers collate all the data into a dictionary immediately after they calculate it. Lots of hashtable lookups to make it work. Do I need to write some mutex stuff or does multiprocessing handle mutex?