Just a side issue, but an important one. I think it was Jeff who some time ago posted something about always validating any string being converted to a datetime datatype....

Just for fun I decided to run each of these three methods (plus a fourth one I thought of later) to see which would perform the best. It's hard to...

You're correct. An oversight...

Sean Lange (2/8/2013)

---

dwain.c (2/7/2013)

---

How about something like this to avoid the self-join? Uses Sean's setup data.

SELECT Div_time, Unit_id

,TimeDifferenceMin=DATEDIFF(minute, MIN(Filled_Time), MAX(Filled_Time))

FROM #Something

GROUP BY Unit_id, Div_time

HAVING MAX(Filled_Time)...

--create some test data

IF OBJECT_ID('tempdb..#TempTable') IS NOT NULL

DROP TABLE #TempTable

CREATE TABLE #TempTable (

[ComputerID] INT IDENTITY(1,1) NOT NULL,

[ComputerName] NVARCHAR(50) NULL,

...

This isn't too different than what Scott offered but here's my contribution:

LOCATION

LocationID

LocationType

LocationParent (let's you build a hierarchy)

LocationName

(etc)

LOCATION_TYPE

LocationTypeID

LocationTypeDesc (content examples: HQ, Store, Warehouse, Fulfillment, etc)

CONTACT

ContactID

LocationID

ContactName

ContactAddress

ContactSort ...

I made an assumption that you make two readings and two readings only on each machine. That assumption let's me join the table on itself and put the two readings...

JP10 - Here's a timely article on table partitioning posted in today's SQL Server Performance Newsletter:

http://www.brentozar.com/archive/2013/01/best-practices-table-partitioning-merging-boundary-points/

 

Sounds like you have two problems. Large inserts and even larger deletes. I'd do something like this:

1. First create a partitioned table.

MSDN: http://msdn.microsoft.com/en-us/library/ms188730(v=sql.105).aspx

2. Import the 4000 inserts into a heap...

SQLJocky (2/6/2013)

---

Thanks for the response. Yes, the company has a DR site already setup with VMWare infrastructure and are planning for everything. We just need to determine what...

Full-text search is great but you must change your thinking a bit. It is a ranking system based on proximity (how close together are the search terms) and inflection (grammatical...

SQLJocky (2/6/2013)

---

We have a very large client that will have 5000+ stores that will be connecting in on a web app and I'm looking for the best architectural design to...

Then one option would be to create the export via Tasks | Export Data and save it as a package you...

It's not SSIS, but from SSMS you can right-click on the database, select Tasks | Generate Scripts. This utility will let you select whatever objects you want to copy including...

ChrisM@Work (2/5/2013)

---

Steven Willis (2/5/2013)

---

Just a generic observation...

If this function is taking two hours to run--even with all of the cursors and subcursors--then you are probably accessing a LOT of data....