We're looking at optimizing some of our tables because we have indexes on columns that are not used. So for example we might have a table that has 6GB of data and 4GB in indexes (according to sp_spaceused). We need to know how much of the 4GB of indexes is consumed by each of the indexes individually. I've tried dbcc showcontig, but that doesn't tell me the amount of space used by the index that I run it for.
Second (and by far not as important), is there a way to determine how much space is being consumed by data in a specific column. The original creator of some of the tables added an incrementing identity column that is used no where in the system. I'd like to be able to say..."If we drop this column we'll save XXX in space."
Knowing the index space is more critical, the column space would be nice but not necessary.
use tempdb go if object_id('Data', 'u') is not null drop table Data go with temp as ( select top 10000 row_number() over (order by c1.object_id) Id from sys.columns c1 cross join sys.columns c2
[code]....
What index would be best for these three queries? With best I mean the execution time, I don't care about additional space.
This is the index I currently use: create nonclustered index Ix_Data on Data (StateId, PalletId, BoxId, Id) The execution plan is SELECT (0%) - Stream Aggregate (10%) - Index Scan (90%).
Can this be optimized (maybe to use Index Seek method)?
I'm trying to establish the mb usage of a series of nonclustered indexes, I'm used to using the manage indexes GUI in 6.5, and showcontig doesn't quite give me what I want, any suggestions?
i am trying to make a highscore table and i would like when the user enters a score for them to be able to see what position they are in the database without having to take out all the data. i have looked at a few methods and was wondering which one is the best solution and how i would go about implementing it
I want to write SQL that will search the tables in a database for a specific column, like this. For instance, I have a column "Unique_ID" that is in many of our tables (hundreds) but not in others and want to find out the tables it is in. It is always the first column.
I tried to find a system stored procdure to do this but couldn't and tried to create a script using the sysobjects and syscolumns tables in the Master db, but came to a roadblock because they don't seem to be related at all.
I would surely appreciate if someone else has already done this!
SELECT CASE WHEN CHARINDEX(' ', Supervisor, CHARINDEX(' ', Supervisor, 0) + 1) > 0 THEN REPLACE(Supervisor, SUBSTRING(Supervisor, CHARINDEX(' ', Supervisor, CHARINDEX(' ', Supervisor, 0) + 1), LEN(Supervisor)), '') ELSE Supervisor END AS NewSupervisor from data d
However, I stumble when Middle Name exists somewhere in the name as Replace function repalces every occurrence of the string. For ex: "Allen, Gary G" becomes "Allen,ary"
Do we have any way to say sql to replace after certain index?
Ok, please pardon my complete ignorance; but I am extremely SQL challenged. I manage the server right now and I am the closest thing to a DBA at the moment with this server.
Here is the problem. I have a job that runs every hour and until last week it was running without any problems. Now, once a day it hangs, and I end up having to restart sql to get it to be functional again. I guess I could just kill the spid, but regardless, it's fails.
It doesn't happen at the same time each day. I thought it was from the backups doing a SQL backup job, but I eliminated that possibility. I can't find what is causing the problem, because the SQL logs just stop at the time that this happens. It's like some other job/query is running and taking all the sql server resources. But I can never catch it when it's happening and I have no event logs or SQL logs to go on.
Is there any query that I can run to find what happened around the specific time the job failed to run? I just need to find what process did what when this job stopped working.
Again, I apologize if my lack of SQL knowledge is making me ask dumb questions, but I am really lost.
insert into #temp (Cabstamp,account,Document, origin, debit, credit, datalc) select 'ADM12345',111,'CMP-01','FO',1000,0, '20150110' union select 'ADM12345',112,'CMP-01','FO', 500, 0,'20150110' union select 'ADM12345',6811,'CMP-01','DO',0,1500,'20150110' union
Can you create an UPDATE TRIGGER and use some typeof code to figure out which SP just updated the current table?If not how can i achieve what i want?I tried to run SQL Profiler and i don't understand why i can'tsimply have the Profiler filter events only for the specific database idand the table's object id i chose?What am i doing wrong with SQL Profiler? I was testing thisthrough SQL EM. I had the filters chosen for a specific database idand a specific table's object id, yet when i open another table SQLProfiler captures that information too.Thank you
write a query which retrieves only unique rows excluding some columns.
IdStatusmanager Team Comments Proj number Date 19391New XUnassigned One 3732.0 16-Apr-14 19392Can YCustomer Two 3732.0 17-Apr-14 19393Can YCustomer Two 3732.0 17-Apr-14 19394Can YCustomer One 3732.0 18-Apr-14 19395New YCustomer One 3732.0 19-Apr-14 19396New YCustomer One 3732.0 21-Apr-14 19397New ZCustomer One 3732.0 20-Apr-14
In the above table project number and id shouldn't be considered and I should get the unique rows considering rest of columns and sorted based on date. Expected result is
IdStatusmanager Team Comments Proj number Date 19391New XUnassigned One 3732.0 16-Apr-14 19392Can YCustomer Two 3732.0 17-Apr-14 19394Can YCustomer One 3732.0 18-Apr-14 19395New YCustomer One 3732.0 19-Apr-14 19397New ZCustomer One 3732.0 20-Apr-14 19396New YCustomer One 3732.0 21-Apr-14
Hi all, We have a table that is full text enabled and it is working fine, but the full text search doesn't returns any record for the following case
select * from let_catalog_search where contains(search_field,'"Bulk Process 1*"') even though there exist records that satisfy the condition in the table, the record that i am talking abt is "bulk process 1 with price bp100-ilt1-00200136 bp100-ilt1"
If I remove the last 1 from the search string i get lot of records, Can anybody help me out.
I have been reviewing the primary keys that our vendor-purchased application has provided. There are no foreign keys and I am reviwing the clusered/nonclustered indices to see whee foreign keys may be needed. I have noticed some clustered indices contain 3 or 4 varchar(20) columns.
1)Is there a maximum size for an index where its size may be counterproductive and perhaps the index slows down processing?
2) Does any one know of software tools that analyzes primary keys and designates possible foreign key candidates?
Are there any references that help with estimating the size of a FTS index, or the whole catalog? I'm trying to do some capacity planning, and can't find anything to help me guess how much disk I'll need.
Hi,How do I find the max row size for a particular table?This was the error I recieved while execting my proc with the relevanti/p I need to:"cannot sort a row of size 8192, which is greater than the allowablemaximum of 8094"I also understand that the max bytesize of a row is 8060 bytes.Butwhtz this 8094?TIA,Seetha
Hi all, Is it possible to get the name and size of each index in a table? Please let me know. In sql 7, we could do this using EM but, in sql 2000, I'm not sure how to do this.
question, since yesterday i have a problem with my indexes, they started growing very big, this morning it stopped growing but now my database is 22 GB instead of 15 GB, i already tried to rebuild my indexes and even tried to recreate my indexes but they didn't get any smaller.
Is somebody formiliar with this problem, the app developer only changed a few tables and they say that it's have nothing to do with it, i don't know what they changed, i'll tell that later.
tia Benjamin
Need an SQLDB consultant? check www.veeningsengineering.nl
I'm trying to find a specific string (a name) and replace it with another inside of a VARCHAR(7000) field. Unfortunately, there are names like Ted and Ken that I'm trying to replace. I would like to leave words like Broken, admitted, etc... intact.
UPDATEtbl SETBody = LEFT(REPLACE(tbl.Body, pm.OldFirstName, p.FirstName), 7000) FROM Table tbl JOIN Person p ON p.PersonID = tbl.PersonID JOIN PersonMap pm ON pm.PersonID = p.PersonID AND LEN(pm.OldFirstName) > 2 WHEREtbl.Body LIKE '%[^a-z]'+pm.OldFirstName+'[., ]%
'The problem I'm running into is that the '[, ]%' in the LIKE excludes any record that ends with the FirstName because it is requiring either a space, comma or period after the name. Is there some way to add an empty string to the list of acceptable characters as that would cover any scenario in the data? I would prefer not to add all characters except space, comma and period, but I guess I could do that.
Env: SQL Server 2000The following sql stmt seems to find a particular table's sizeprogrammatically:select top 1 [rows],rowcntfrom sysindexeswhere ID = object_id('aUserTable')and status = 0and used > 0However,a) I'm not 100% sure of its consistency;b) Both [rows] col and [rowcnt] col seems to produce same data, whichone is supposed to be more accurate (or more up to date)?TIA.
every time i am doing my database shrinking manually.
what i am asking i find out every time when i shrink my database that sql is showing minimum size of database to shrink.how can its getting minimum size
but i need to find out how to get minimum size of particular database programatically.
I'm trying to determine how much space some tables use (SQL2000), and I found 2 suggestions posted earlier, but can't get them to work for me.
The first was "Right click on the DB in Enterprise Manager, select view then taskpad."
When I try that, I get some of the tables displayed with the info I want, but I can't see them all and can't scroll down beyond the first 22 tables in the database.
It says "command copleted successfully, but where does the output to this go ?? Is there something other than "print" I should use ? The grid pane is empty.
hi,There is facility to do backup of the database. I can use thatfacility to backup the database. But before i do a backup i want tocheck if the disk space available enough to backup that database. Ihave a 22gb database mdf file, when i took a backup of that its only3gb. SO i cannot use the size of mdf file to determine the databasedump file. Is there any facility available to find out the backupdatabase dump size before doing the backup?ThanksNabhonil.
Is there any way to get size of the individual column in a table?
I know we can use sp_spaceused to get the size of the table. But my question is diiferent. I have a table with 50 columns and approx 2 million rows in it. I wanted to know which column is taking most of the space.
