Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
sql777
Constraint Violating Yak Guru
314 Posts |
 Posted - 2005-07-27 : 21:29:51
|
| I was reading about some website that gets 2 million new users per month.how would you go about designing such a users table that wouldn't take ages for simply username/password lookups etc. (first name, lastname, email etc).would you create 26 user tables, and add each user to a table depending on the 1st letter of their name --> to spread the users out so table lookup performance won't be degraded.what other ideas are there for such a high load application? |
|
|
nr
SQLTeam MVY
12543 Posts |
Posted - 2005-07-27 : 22:08:37
|
| Heard of indexes? Have a look in bol.==========================================Cursors are useful if you don't know sql.DTS can be used in a similar way.Beer is not cold and it isn't fizzy. |
 |
|
|
sql777
Constraint Violating Yak Guru
314 Posts |
Posted - 2005-07-27 : 23:40:59
|
| Yes I have, but a 20 million row users table will slow down the db bigtime. |
|
|
|
sql777
Constraint Violating Yak Guru
314 Posts |
Posted - 2005-07-27 : 23:41:15
|
| each year that is 24 million new rows btw. |
|
|
|
Michael Valentine Jones
Yak DBA Kernel (pronounced Colonel)
7020 Posts |
Posted - 2005-07-28 : 07:31:05
|
If you have good indexes on the table, the lookups will be fast. 20 million rows is not that big. I have a database where we add several million rows each day, and we have fast lookups by indexed columns.quote:
Originally posted by sql777
Yes I have, but a 20 million row users table will slow down the db bigtime.
CODO ERGO SUM |
|
|
|
Stalker
Yak Posting Veteran
80 Posts |
Posted - 2005-07-28 : 08:03:36
|
| you can also use partitioned views - when one big table is divided in several smaller tables and partitioned view just is based on these tables, it can significally improve perfomance |
|
|
|
|
|
|
just some questions on high traffic, high load