From: <to...@wa...> - 2009-10-23 08:22:38
|
BODY { font-family:Arial, Helvetica, sans-serif;font-size:12px; }Hi Varun, I'm forwarding this back to the lucid list, as when I hit reply to you I forgot to send my reply there, and those guys know more than me, but here's my guesses..... 1) My understanding is that largest the improvement is gained over large dataset where the query on a conventional db would run out of memory 2) Erm, sure. 3) Well if you're importing surely the tables are specifically created for luciddb? 4) Dunno, but I would also agree with that, maybe this can be improved upon. I also come from a MySQL/Postgres background and I know the learning curve is pretty steep, but worth it in the end. And I also know that the MySQL converts are the type of people Lucid need so we need to work out where the problems lie for people like yourself! Cheers Tom On Fri 23/10/09 8:32 AM , Varun Gopinath var...@tr... sent: Hey Tom, i would also like ur opinion on some other things.Would be glad if you could throw some light on the pointers below:- I am basically planning to move my MySQL database to LucidDB. i had a few queries related to it:- 1) What difference in performance can i expect if i move it to LucidDB?2) i have a star schema already in place in Mysql.Now if i transfer the tables using a foreign wrapper to LucidDB, would there be any change in which the schema would be stored? As in, would it still be placed as a star schema? 3) Also, would the compression and storage be more optimised if i create the tables specifically in LucidDB instead of importing the same from some other database? 4) Lastly, i was going through the explain parameters for the ucidDB query, and it seemed so complicated to interpret? is there some way to figure things out or some easier way to understand and interpret slow queries? Thanks in advance for your time.really appreciate it. its really very important for me. Tc Cheers,Varun On Wed, Oct 21, 2009 at 10:55 AM, Varun Gopinath wrote: Thanks a lot buddy. :) you saved so much of my time. On Tue, Oct 20, 2009 at 9:23 PM, wrote: hi Varun By default lucid will change all table names field names etc to upper case unless you surround them in quotes. Try that.. Tom On Tue 20/10/09 1:22 PM , Varun Gopinath var...@tr... [3] sent: hi, I am new to LucidDB and was thinking of moving my database from Mysql to LucidDB due to the advanced efficiency LucidDB provides.I am currently stuck in a situation and would be glad if you could help me out. I am actually trying to migrate data from MySQL server using the SQL data connector 5.1.the statements i have used are as follows:- CREATE or replace SERVER mysql FOREIGN DATA WRAPPER sys_jdbcOPTIONS ( driver_class 'com.mysql.jdbc.Driver', url 'jdbc:mysql://localhost/test_data?useCursorFetch=true', user_name 'root', password '', login_timeout '10', validation_query 'select 1', fetch_size '10000', table_types 'TABLE', schema_name 'TESTING'); After this, i created a schema named 'Extraction_Schema' and did the following import. import foreign schema TESTING from server mysql_testinto extraction_schema; Now when i run the query " select * from sys_root.dba_stored_tables;", i get the output which indicates as follows:- | LOCALDB | EXTRACTION_SCHEMA | timetable | | LOCALDB | EXTRACTION_SCHEMA | employee | | LOCALDB | EXTRACTION_SCHEMA | user | This clearly shows that the data has been populated in the schema. Now when i run the following query:- set schema 'Extraction_Schema' select count(*) from employee; it gives me the following error:- Error: From line 1, column 15 to line 1, column 22: Table 'EMPLOYEE' not found (state=,code=0) Could somebody please let me know where and what is going wrong?I would really appreciate some help in this regard. -- Regards, Varun Gopinath "To be the best, you need to survive the worst" -- Regards, Varun Gopinath "To be the best, you need to survive the worst" -- Regards, Varun Gopinath "To be the best, you need to survive the worst" Links: ------ [1] mailto:var...@tr... [2] mailto:to...@wa... [3] mailto:var...@tr... |