[Progress Communities] [Progress OpenEdge ABL] Forum Post: RE: A hundred million rows?

Status
Not open for further replies.
D

dbeavon

Guest
>> When you say "via a DDL statement" are you implying that you are using SQL to add the index? Yes. For the majority of our database schema it is possible to maintain it via SQL statements. But it seems as if there is a limit that you eventually reach based on the size of data, where you have to switch modes and start using proutil commands as part of the workflow. It feels like "cheating"... mainly since proutil doesn't feel like a regular part of the DBMS, but more like a back-door tool for editing the internal files *without* the help of a DBMS layer. (Perhaps it would at least feel less like "cheating" if this stuff could be accomplished with the standard data dictionary user interface). >> If I had a 10 hour outage window to work with I would add the index inactive and do a simple offline proutil idxbuild. Yes, I think I'll have to start factoring in the need for proutil whenever working with our larger OE tables. I guess the SQL/DDL interface isn't cut out for that. >> PUB.table 24,987,814,259 Wow, I had to put the commas in there to see that this is table that is 250 times more difficult to manage than the one I'm dealing with. I'm betting that is a table whose schema is rarely changed. Maybe in its own database? Full backup only once a year? What happens if a user kicks off the wrong query? It hurts to think about...

Continue reading...
 
Status
Not open for further replies.
Top