1,631 reputation
724
bio website
location Austin, TX
age 33
visits member for 3 years, 8 months
seen 14 hours ago

18h
comment changing SQL data types from varchar(50) to smaller rational types increases table space
If you'd like to compact the table where it stands, it looks like this is covered on SO: stackoverflow.com/a/808368/565869.
18h
comment changing SQL data types from varchar(50) to smaller rational types increases table space
Did you import the data into table A, then ALTER TABLE to make it into table B? Try creating table B from scratch and then importing the data into it.
Jan
23
comment Physical Address Schema
I don't know how this could work; a ZIP code does not uniquely identify a street address. Perhaps Portuguese postal codes do, but ZIP codes do not.
Jan
23
comment Best engine type for WHOIS data
What exactly do you want to store? Just { DomainName, Name, DateRegistered }, or boatloads of details for each? If your concern is that a WHOIS entry can have many contacts, and not the same set for each, you're dealing with something similar to addresses in a contact database (dba.stackexchange.com/questions/9256/physical-address-schema).
Jan
21
comment How will large index INCLUDE fields effect system performance?
How long are the actual values? A VARCHAR(2000) which typically stores just ten characters is one thing; a solid 2,000 bytes per record is something else.
Jan
21
comment Covering Index - Insert Performance
What exactly is your question? "How do I find the tradeoff?" isn't very clear. If you're trying to decide whether to keep the covering index, it depends on how important inserts are relative to queries. I don't see how your covering indices on D or E are helping, as they duplicate the entire table. What do your queries look like?
Jan
20
comment Constellation Schema; Relationships between fact tables
Can you tell us something about the relationship you're considering? I've had fact-to-fact joins in star schemas before, but only as a last resort.
Jan
20
comment SQL Server - Slow Database for no apparent reason
How much data is in each system? If your test environment has 1% the records, then much longer run times in prod shouldn't be surprising. Also, are all, or most, or only a few queries much slower in prod?
Jan
20
comment How to avoid using Merge query when upserting multiple data using xml parameter?
The issues with MERGE that Bertrand points out are mostly edge cases and inefficiencies, not show stoppers - MS wouldn't have released it if it was a real minefield. Are you sure that the convolutions you're going through to avoid MERGE are not creating more potential error than they're saving?
Jan
20
comment Why is it a bad practice to allow everyone to use the sa login?
There is an upside, it's just outweighed by the others: convenience. It's very easy for everyone to use sa (probably with "password" as the password), that's why it continues as a practice.
Jan
20
answered Datawarehouse bad practice?
Jan
14
comment What's the best practice to avoid collisions in throwaway temp tables?
In this particular case, the temp table was to store aggregated data with a defined key, so I could use it in a multi-table UPDATE. Good stuff in that linked answer, thanks!
Jan
14
asked What's the best practice to avoid collisions in throwaway temp tables?
Jan
14
answered Is there a term for a table that has an “effective date” column?
Jan
9
awarded  Organizer
Jan
9
revised Speed inserts in database with a table with +47 million data per each year, containing more than 10 years?
The previous tags did not apply at all
Jan
9
comment What is the cause of this problem with CONVERT()?
@AaronBertrand: Would you like to make that an answer? We can mark this off the "unanswered" list.
Jan
9
suggested approved edit on Speed inserts in database with a table with +47 million data per each year, containing more than 10 years?
Jan
9
comment Get all the data between Characters
There are several questions on DBA and StackExchange about splitting concatenated values. My favorite: dba.stackexchange.com/a/21081/1896.
Jan
8
accepted What steps are necessary to safely restart a machine hosting Microsoft SQL Server?