How to solve the problem of creating-closing processes?

Question

Python

There is

the code

from multiprocessing import Pool

def execute(album):
   pool = Pool(processes=8)
   result = pool.map(execute_the_photo, album)
   pool.close()
   pool.join()
   *save to file*

for album in albums:
   execute(album)

It turns out that every time the first aka main cycle is run, processes are created and closed.
It takes too much time to create.
What are the options in which processes are created once at the very beginning of processing, then they constantly calculate the data, and at the end of the work, they are closed?

Answer 1

Google: "ProcessPoolExecutor python"
You create a pool at the beginning of work and it's done

Answer 2

Try prefork
www.michurin.net/servers/prefork-server.html
In general, there are several options for the server
1 thread
2 fork (what you have)
3 prefork (thread is created before use, dynamic control of the number of processes is possible)
4 thred
5 async