Hi All,
I need an advice, take a look at this code sample:
engine = create_engine(url,encoding = 'utf-8',echo = False,pool_size =
100)
metadata = MetaData()
object_table = Table('_object', metadata,
Column('id', Integer, primary_key=True),
Column('_owner_id', Integer),
Column('_created', DateTime),
Column('_creator_id', Integer),
Column('_modified', DateTime),
Column('_modifier_id', Integer),
Column('class_id', Unicode(60)),
Column('type', Integer),
Column('singular_label', Unicode(200)),
Column('plural_label', Unicode(200))
)
class Object(object):
pass
mapper(Object,object_table)
def application(environ, start_response):
Session = sessionmaker(bind=engine)
session = Session()
status = '200 OK'
"""
#1. use session
output = '\n,'.join(['%s: %s %s %s %s %s %s %s %s'%(o.id,
o._owner_id,o._created,o._created,o._creator_id,
o.class_id,o.type,o.singular_label,o.plural_label)
for o in session.query(Object)])
"""
#2. use raw sql
s = text('''
SELECT _object.id AS _object_id,
_object._owner_id AS _object__owner_id,
_object._created AS _object__created,
_object._creator_id AS _object__creator_id,
_object._modified AS _object__modified,
_object._modifier_id AS _object__modifier_id,
_object.class_id AS _object_class_id,
_object.type AS _object_type,
_object.singular_label AS
_object_singular_label,
_object.plural_label AS _object_plural_label
FROM _object
''')
output = '\n,'.join([str(p) for p in session.connection().execute
(s).fetchall()])
session.close()
with open('/home/alex/integrity/data/a1/xdoc/schema.xml','r') as
f:
output = '%s %s'%(f.read(),output.encode('utf-8'))
response_headers = [('Content-type', 'text/plain'),
('Content-Length', str(len(output)))]
start_response(status, response_headers)
return [output]
As you see there are 2 options for query - raw mysql query and ORM
based one.
I've performed some tests on my PC
ab -n 1000 -c 100 http://localhost/l
ORM: (option 1 turned on)
Document Path: /l
Document Length: 87305 bytes
Concurrency Level: 100
Time taken for tests: 17.352274 seconds
Complete requests: 1000
Failed requests: 0
Write errors: 0
Total transferred: 88662825 bytes
HTML transferred: 88439965 bytes
Requests per second: 57.63 [#/sec] (mean)
Time per request: 1735.227 [ms] (mean)
Time per request: 17.352 [ms] (mean, across all concurrent
requests)
Transfer rate: 4989.78 [Kbytes/sec] received
RAW: (option 2)
Document Path: /l
Document Length: 89533 bytes
Concurrency Level: 100
Time taken for tests: 8.951501 seconds
Complete requests: 1000
Failed requests: 0
Write errors: 0
Total transferred: 89753000 bytes
HTML transferred: 89533000 bytes
Requests per second: 111.71 [#/sec] (mean)
Time per request: 895.150 [ms] (mean)
Time per request: 8.952 [ms] (mean, across all concurrent
requests)
Transfer rate: 9791.54 [Kbytes/sec] received
Pentium 4 2.4/1.5GB mem/apache2(mpm-worker)/mod_wsgi
I've done lot's of attempts to avoid occasions and formula is the
same: performance drops almost 2 times. And this distance grows when
increasing number of iteration.
My "big" app issues about 5 queries using ORM, so in the end I get
10-13 requests/second (ORM) vs 35-40 requests/second using raw.
What can be done to increase the performance of my server in ORM based
case?
Any advice would be appreciated.
Regards, Alex
P.S.
this part:
with open('/home/alex/integrity/data/a1/xdoc/schema.xml','r') as f:
output = '%s %s'%(f.read(),output.encode('utf-8'))
is here just to add some traffic to response (file size is 7kb )
There are ~20 records in the _object table
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"sqlalchemy" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/sqlalchemy?hl=en
-~----------~----~----~----~------~----~------~--~---
