Why does Django ORM not have time to update the connector in multithreading on HDD?

Question

Django

I ran into a problem that on HDD Django ORM does not have time to update the connector in a parallel project in which I connected the connection to the database through Django ORM

import django
SITE_PATH = os.path.abspath(os.path.dirname(__file__))
PROJECT_PATH = os.path.normpath(os.path.join(SITE_PATH, '..'))
SRC_PATH = os.path.join(PROJECT_PATH, 'api')
if SRC_PATH not in sys.path:
    sys.path.insert(0, SRC_PATH)

os.environ.setdefault("DJANGO_SETTINGS_MODULE", "api.settings")
django.setup()

connected the models of the required table via from import and passed the table to several rabbitMQ queues.
When I run it on hardware with an SSD, it runs normally in 5 threads, but on hardware with an HDD, it normally works only in 1 thread, and if there are already 2 or more, it starts to fall off. There is a feeling that the connector does not have time to update and I get data retrieval errors (

Why can this be? Can someone explain?

I want to try using SQLAlchemy to connect to this database and access it in parallel with the Django project