def my_func(df):
#mach iwas mit dem df
return df

dfs = func_to_generate_df_chunks(df) #<- liste mit deinen chunks die concurrent laufen sollen

#MultiProcessing:
from concurrent.futures import ProcessPoolExecutor
num_processes = 40
with ProcessPoolExecutor(40) as pool:
	#liste mit processes:
	futures = [pool.submit(my_func, df) for df in dfs]
	result = pd.concat([future.result() for future in futures])