I'd say: with the propper DB and DBI (PostgrSQL + Perl DBI, ie), using
PREPARE and COMMIT, SQLBox is your best bet by far.

Regards,


Alberto Mijares




On Wed, Apr 22, 2015 at 12:50 AM, Makhanu Sinja <jeysi...@gmail.com> wrote:
> Well last week The same issue was raised on another thread in this
> mailing list. Let us consider the amount of time sqlbox has to do
> database CRUD for 1M messages compared to using spool or files. Is
> there anyone who has worked with both?
>
> On 4/21/15, Rene Kluwen <rene.klu...@chimit.nl> wrote:
>> 1.
>> I think Kannel does support Keep-Alive connections. Not sure about the
>> server side, but I think it does. Just make sure your client also supports
>> it.
>>
>> 2.
>> Yes, probably SQL Box does insert 1,000,000 a lot faster than you can do by
>> http. Question is if your upstream providers handle such a rate. You will
>> end up with a lot of pending messages in the bearerbox queue.
>>
>> == Rene
>>
>>
>> -----Original Message-----
>> From: users [mailto:users-boun...@kannel.org] On Behalf Of Cliff Court
>> Sent: dinsdag 21 april 2015 10:11
>> To: users@kannel.org
>> Subject: Fastest method to insert a million MT messages to Kannel
>>
>> Hi All
>>
>> I have set up Kannel with bearerbox and smsbox, which is working and I am
>> writing dlr's to a mysql db.
>>
>> Currently I am submitting messages using a sendsms GET or xml-based POST, as
>> per the Kannel documentation. However, using individual http GET or POSTs
>> for each message is relatively slow when needing to send a large volume of
>> messages. Let's assume I have a 100 msgs/sec connection to an external SMSC
>> using SMPP from bearerbox, which will take ~3 hours to send 1 million MT
>> messages.
>>
>> So my question is what is the fastest method to submit messages to bearerbox
>> to send a million MT messages?
>>
>> I have seen that SQLbox is available and that one can insert messages into
>> the send_sms table for faster submission, but I'm wondering what is the
>> fastest method of submission to bearerbox?
>>
>> Thanks
>> Cliff

Reply via email to