Suppose we have multiple sites (using sites-framework of Django) running on the same django instance.
MyModel.objects.filter(site_id=request.site, slug=slug)[0]
might be an overhead in the future. So I was wondering if I could split the databases for this model(s) for querying faster.
So if request.site
is 1, running
MyModel.objects.get(slug=slug)
will query only db1.
If request.site
is 2, running
MyModel.objects.get(slug=slug)
will query only db2.
I will be getting request.site from the request parameter of the view as value of site will be dynamically determined according to the subdomain used: de, fr, etc.
You may create a custom router to do this:
Example:
def get_current_site():
SITE_ID = getattr(settings, 'SITE_ID', 1)
site_name = Site.objects.get(id=SITE_ID)
return site_name
DATABASE_ROUTERS = ['CustomDatabaseRouter',] #a setting that Django understands.
class CustomDatabaseRouter(object):
def db_for_read(self, model, **hints):
site_name = get_current_site()
if site_name in ['site1']:
return 'db1'
if site_name in ['site2']:
return 'db2'
return 'default'
def db_for_write(self, model, **hints):
site_name = get_current_site()
if site_name in ['site1']:
return 'db1'
if site_name in ['site2']:
return 'db2'
return 'default'
def allow_syncdb(self, model, **hints):
site_name = get_current_site()
if site_name in ['site1'] and db == 'db1':
return True
if site_name in ['site2'] and db == 'db2':
return True
return False
You can readup more here
https://docs.djangoproject.com/en/dev/topics/db/multi-db/