[MonetDB-users] writing to MonetDB via ADO
Hi, I am using Microsoft ActiveX Dataobject (ADO) to write my records to the MonetDB database. I am inserting records row by row (running it out of VB environment): set cnnMONETDB = new ADODB.Connection for i = 0 to rownum createsql = "INSERT INTO tablename (columnnames) VALUES (values);" cnnMONETDB.Execute createsql next i It takes me about 5min to insert 100000 rows into the database, which is too slow for my purpose. Is there any way to speed this up? I was reading that monetdb can update 100000r records per second. i must be doing something wrong then :( Thanks and regards Der WEB.DE SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! *http://smartsurfer.web.de/?mc=100071&distributionid=000000000066* [http://smartsurfer.web.de/?mc=100071&distributionid=000000000066]
Hi, Thanks for using MonetDB/SQL. I'm not too familiar with visual basic, but I assume your "(columnnames)" and "(values)" are filled in with actual values by your application. Hence, looking at what you give us, I suspect that you didn't switch off auto-commmit mode. Doing so will improve performance (sometimes dramatically). An even higher speed can be achieved when using an COPY INTO construct, but this is harder to deal with from an application environment like you seem to be in. Regards On 26-10-2006 09:26:15 +0200, baane@web.de wrote:
Hi,
I am using Microsoft ActiveX Dataobject (ADO) to write my records to the MonetDB database. I am inserting records row by row (running it out of VB environment):
set cnnMONETDB = new ADODB.Connection
for i = 0 to rownum
createsql = "INSERT INTO tablename (columnnames) VALUES (values);"
cnnMONETDB.Execute createsql
next i
It takes me about 5min to insert 100000 rows into the database, which is too slow for my purpose. Is there any way to speed this up? I was reading that monetdb can update 100000r records per second. i must be doing something wrong then :(
Thanks and regards
Hello, I've been evaluating OODBMS-options, and I've come across the papers describing MOA. Some dead links google gives me about it contain something about a MOA Frontend to MonetDB, while the only place it's referenced on the website is a link to Tweente-University where I couldn't find anything informative though. I'm building a system that will need to store representations of deep hierarchical structures and MOA might be an interesting option. I couldn't find anything but papers (and tests in src/from_tweente), so my question: is there a working or maybe incomplete, just any implementation of MOA I could evaluate? Or maybe actual implementations do not carry the name MOA? I'd really like to use MonetDB for this. Especially Version Five+MAL has features like coroutines, boxed variables and factories which might get very interesting for my project (once I'm comfortble with the basics/version 4 ). If there's no implementation of moa, maybe someone has an alternative suggestion on how to represent hierarchical data in MonetDB? Thanks for consideration, Johann
Hello, I've been evaluating OODBMS-options, and I've come across the papers describing MOA. Some dead links google gives me about it contain something about a MOA Frontend to MonetDB, while the only place it's referenced on the website is a link to Tweente-University where I couldn't find anything informative though. I'm building a system that will need to store representations of deep hierarchical structures and MOA might be an interesting option. I couldn't find anything but papers (and tests in src/from_tweente), so my question: is there a working or maybe incomplete, just any implementation of MOA I could evaluate? Or maybe actual implementations do not carry the name MOA? I'd really like to use MonetDB for this. Especially Version Five+MAL has features like coroutines, boxed variables and factories which might get very interesting for my project (once I'm comfortble with the basics/version 4 ). If there's no implementation of moa, maybe someone has an alternative suggestion on how to represent hierarchical data in MonetDB? Thanks for consideration, Johann
participants (3)
-
baane@web.de
-
Fabian Groffen
-
Johann Borck