Oninit cdi / Archive
Talk to an expert →

Re: Table split into separate server and instance

Posted in 2003

Topics: Performance & Tuning, SQL Development & Query Writing, Stored Procedures & SPL, Security, Permissions & Auditing

kh teoh — — source: Usenet: comp.databases.informix

Michael Krzepkowski <NOmichaelkSPAM@sqlcanada.com> wrote in message news:<3EEA1186.3020204@sqlcanada.com>... > You can split them (create two tables, then populate them) and you can > join them > in one select - however do not expect the performance to be any good if > these tables > are huge. Hint: any column in Informix can be expresses as > "database@instance:tablename.column" > > HTH > > Michael Is it like this ? SELECT HR@instance1:employee.id, HR@instance1:employee.name, HR@instance1:employee.address, HR@instance1:employee.telno, HR@instance1:employee.dob, HR@instance2:employee.epfno, HR@instance2:employee.password, HR@instance2:employee.salary WHERE HR@instance1:employee.id = HR@instance2:employee.id

↪ replying to kh teoh
Michael Krzepkowski — — source: Usenet: comp.databases.informix

kh teoh wrote: >Michael Krzepkowski <NOmichaelkSPAM@sqlcanada.com> wrote in message news:<3EEA1186.3020204@sqlcanada.com>... > > >>You can split them (create two tables, then populate them) and you can >>join them >>in one select - however do not expect the performance to be any good if >>these tables >>are huge. Hint: any column in Informix can be expresses as >>"database@instance:tablename.column" >> >>HTH >> >>Michael >> >> > >Is it like this ? > >SELECT HR@instance1:employee.id, HR@instance1:employee.name, >HR@instance1:employee.address, HR@instance1:employee.telno, >HR@instance1:employee.dob, HR@instance2:employee.epfno, >HR@instance2:employee.password, HR@instance2:employee.salary >WHERE HR@instance1:employee.id = HR@instance2:employee.id > > Yes.

↪ replying to kh teoh
Jonathan Leffler — — source: Usenet: comp.databases.informix

kh teoh wrote: > Michael Krzepkowski <michaelk@sqlcanada.com> wrote: >> You can split them (create two tables, then populate them) and >> you can join them in one select - however do not expect the >> performance to be any good if these tables are huge. Hint: any >> column in Informix can be expresses as >> "database@instance:tablename.column" > > Is it like this ? > > SELECT HR@instance1:employee.id, HR@instance1:employee.name, > HR@instance1:employee.address, HR@instance1:employee.telno, > HR@instance1:employee.dob, HR@instance2:employee.epfno, > HR@instance2:employee.password, HR@instance2:employee.salary > WHERE HR@instance1:employee.id = HR@instance2:employee.id More or less - it depends on how much typing you like to do, and whether you care about the syntax error :-) SELECT E1.id, E1.name, E1.address, E1.telno, E1.dob, E2.epfno, E2.password, E2.salary FROM HR@instance1:employee AS E1, HR@instance2:employe AS E2 WHERE E1.id = E2.id; That add's the missing FROM clause (syntax error), and uses table aliases to make it easier to spot which material comes from which database. -- Jonathan Leffler #include <disclaimer.h> Email: jleffler@earthlink.net, jleffler@us.ibm.com Guardian of DBD::Informix v2003.04 -- http://dbi.perl.org/

↪ replying to Michael Krzepkowski
kh teoh — — source: Usenet: comp.databases.informix

I've got error when I do this :- select a.branch_code, a.user_id, b.branch_code, b.user_id from db4@TEST:code_pwdaccess a, db6@TEST:code_pwdaccess b# ^ # 908: Attempt to connect to database server (test, conerr=-25555, oserr=0 fail ed. # where a.user_id = b.user_id What I did wrong here ?

Related threads