I only suggest that because usually making a connection takes next to no
time at all. I've written my apps different ways - one way opening a
connection and keeping it open and the other way, opening the connection
and closing it on each query. It makes no difference really as far as speed
is concerned, but when i create the connections as i need them, I can run
asynchronously.
On Wednesday, May 29, 2013 5:38:00 PM UTC-7, Derek wrote:
>
> Have you tried to enable caching?
>
> On Sunday, May 26, 2013 1:07:53 PM UTC-7, Saba wrote:
>>
>> Thanks for the reply. I really appreciate. I'm developing an app using
>> web2py to pull up records from mysql db and present it to the user using
>> custom views. It looks like sometimes it takes < 1sec to display the
>> records to the user (i'm the only user now) but other times it takes around
>> 4-5secs to display the same records (less than 50 rows). From my
>> observation, the latency seems to occur when the user, after pulling up
>> records, remains idle for few mins and then pulls up the same records
>> again. I'm using a pool_size of 5. Any suggestions on how to reduce the
>> latency? Again I really appreciate the responses.
>>
>>
>> On Sunday, May 26, 2013 6:37:19 AM UTC-7, Massimo Di Pierro wrote:
>>
>>>
>>>
>>> On Sunday, 26 May 2013 01:34:40 UTC-5, Saba wrote:
>>>>
>>>>
>>>> Hi,
>>>>
>>>> I have a question please regarding db connections and
>>>> connection pooling in web2py.
>>>>
>>>> In models/db.py, I have a connect string like this:
>>>>
>>>> db=DAL('mysql...',pool_size=5)
>>>>
>>>> does model files get executed for every request?
>>>>
>>>
>>> yes
>>>
>>>
>>>> If so, then does it create the above db connection for every request?
>>>>
>>>
>>> No. when the request is completed the connection is put in a pool and
>>> the next request pulls it from the pool. If you have more concurrent
>>> requests than the pool size then a new connection is created. Pool size
>>> starts at zero and grows up to the pool size you have declared (5) than
>>> stay fixed.
>>>
>>>
>>>>
>>>> For eg, if there are 10 requests, does it create 50 separate db
>>>> connections ( 10 requests * 5 pool_size)?
>>>>
>>>
>>> No. If you have 10 requests but not concurrent, it creates only 1
>>> connection in total. If you have 10 concurrent requests, it creates 10
>>> connections (one each) but when complete will close 5 and recycle 5 for
>>> future use.
>>>
>>>
>>>>
>>>> Please let me know. Appreciate your response.
>>>>
>>>> thanks,
>>>> Saba
>>>>
>>>>
>>>>
>>>
--
---
You received this message because you are subscribed to the Google Groups
"web2py-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
