[Articles] The Five Year Plan |
| Posted: 12 Jun 2013 11:00 PM PDT Is your IT department outdated? There's a prediction that they won't be around in five years. Steve Jones doesn't agree. |
| You are subscribed to email updates from SQLServerCentral.com Articles tagged Editorial To stop receiving these emails, you may unsubscribe now. | Email delivery powered by Google |
| Google Inc., 20 West Kinzie, Chicago IL USA 60610 | |
My company is thinking of implementing Change Data Capture. I find many online articles, which cover the basics of Change Data Capture, but no real-life examples of using it, the pros and cons, limitations or "gotchas" encountered during actual implementation and usage, etc. I could use such info as we determine whether Change Data Capture is the right solution for our needs.
Some details about our scenario:
1. Source data: We have currently about 200 entities that use one of our applications. The entities involved are all functioning units of a larger, single business entity, but are spread out geographically. Each entity and its copy of the application stores its data in a separate database, although all copies of the database have identical data structures and tables. They post summarative data on a routine basis to a couple of main servers, however, in each case this data is still stored in separate database on the servers - thus, still about 200 databases.
2. What we would like to do: Bring together this summarative data into a single relational data store, and from there into a dimensional database. We do not want to, nor can we, eliminate the source databases, on either the geographically-distributed sites or the main servers, b/c we don't "own" them.
Is this a feasible scenario for Change Data Capture? We would need to include data column(s) to indicate data sources b/c the data, although placed in a single database, must still be identifiable by data source and data from one source never overwrites data from a different source. As I understand it, Change Data Capture would need to be enabled on the 200+ databases on the main servers, from which we would then likely use SSIS to process the data to post it to the new database we would develop.
I would appreciate any advice, based on understanding and experience, of important things to consider, things to avoid, "gotchas", best practices, etc. We have worked quite a bit with SSIS, developing dimensional data structures, etc., so although any major comments there are welcomed, our primary "need to know" is how/if Change Data Capture can be used for our scenario, as we have no experience with it.
Thanks for your assistance,
Randy
social.technet.microsoft.com/Forums
Microsoft is conducting an online survey to understand your opinion of the Technet Web site. If you choose to participate, the online survey will be presented to you when you leave the Technet Web site.
Would you like to participate?
social.technet.microsoft.com/Forums
We want to delete data (switch partitions) older than 2006 from a partitioned table. There is 2003, 2004, 2005 data that we want to switch, this comes to about 1.5 Billion rows.
While working on switching partitions, found out that up until 20071229, all the data is in one partition,
select @partition.TF_TBL_Fact_Store_Satles_trans(20071229) is 1. and switching partition, will switchout data up until this point. I need to retain 2006 and 2007 data.
What is the best approach to this?
2006 and 2007 data is about 1.05 billion rows and older than 2006 is about 1.4 billiion rows(need to get rid of this 1.4B data and keep 2006/7).
How best can I handle this situation? Any inputs is much appreciated.
Thanks,
Suresh.
Suresh Channamraju
social.technet.microsoft.com/Forums
Weekly SQL Server newsletter with articles, forum posts, and blog posts via email. Subscribers receive our white paper with performance tips for developers.
sqlteam.com
[SQL 2012] Page life expectancy diving to sub-10 on a 128 GB server |
| Page life expectancy diving to sub-10 on a 128 GB server Posted: 27 Feb 2013 09:36 AM PST Last weekend we migrated our primary ERP system from a Windows 2003, SQL 2008R2 server to a windows 2008R2, SQL 2012 server.The new servers are much beefier than the old ones, including an extra 32 gig of ram. The ERP database is ~500 gig, of which a good amount is audit data. On the old server I generally had page life expectancy values around 1000 or more for normal daily activity.On the new server I have been capturing perfmon stats for the past week and I see the PLE jumping all over the place. It climbs up into the several hundred but then suiddenly dives to ridiculously low values like 12 or even 7. The server usage profile has not changed, it's still the same users and applications doing the same things they were doing last week.The instance is set to have min 70 gig, max 110 gig allocated to SQL.The service account has lock pages in memory permission.Can anyone think of some setting I might have missed during migraiton, or some new option to be set in SQL2012 that I've overlooked, that could explain this? |
| Trying to add sharepoint list adapter to VS toolbox Posted: 12 Jun 2013 09:26 PM PDT Hi guys, this msi comes from codeple: https://sqlsrvintegrationsrv.codeplex.com/wikipage?title=SharePoint List Adapters&referringTitle=DocumentationI am trying to add it to my toolbox in 2012. No matter what I do, I have followed the suggestions they give, I cant pull it into my toolbox.Any suggestions?There is an x86 and an x64 version. I have tried both. also both run as admin. no luck. anyone out there with the same issue?thanks a mil |
| Posted: 12 Mar 2013 08:31 PM PDT Hi All,I've been doing a bit of searching round for ways to unit test cubes and the results seem to be limited. What we're trying to do is make sure that any releases we make to live don't alter the existing numbers, unless we are aware that this should be the case, obviously we'd then make a change to the tests.I've been toying with the idea of creating framework using ADOMD.NET, but this appears to be limited to single result set numbers, e.g.[code="sql"]Select [Measures].[Total Sales Value] On 0From [My Cube]Where [Date].[Year - Month - Day].[Date].&[20130313][/code]Does anyone have any ideas? In the meantime I'll continue searching round for ADOMD.NET stuff and post back any results I find.Cheers,Jim. |
| Posted: 12 Jun 2013 09:02 AM PDT In our current system, we query the LDAP using SQL 2008R2 but are moving to SQL 2012 where we have hit a snag. According to [url=http://msdn.microsoft.com/en-us/library/cc707782.aspx]http://msdn.microsoft.com/en-us/library/cc707782.aspx[/url] one of the discontinued features of SQL 2012 is the AD helper service. Does anyone know of a way to query the AD from SQL 2012? |
| Avoiding a database heart attack Posted: 12 Jun 2013 08:25 AM PDT Not long ago, my friend Ayman (DBA/BI developer at the American College of Cardiology in D.C.) and I were discussing the immense value for IT professionals when they observe and interact with their business. Interestingly, Ayman had produced a blog series on the value he reaped from interacting with his business (watching open heart surgery) that we transformed into a seven page guide entitled 'How to Avoid a Database Heart Attack: http://www.infusivesolutions.com/how-to-avoid-a-database-heart-attackNow the reason I choose to post this in a technical forum is this. I recently spoke with SQL MVP Kevin Kline who explained to me that if he could do a 'transaction rollback' and change one thing in his life, it would be to advise his earlier self that IT is not just IT ... it is a means to solve business problems. Consequently, I believe that digesting information explaining how you can reap value from business observation to be of great importance.Nonetheless, for those who don't want to check out the content, I'd like to open a dialogue. How do you interact with the business at your firm and how has it helped you better perform in the IT department? |
| Cores not being used on Windows 2008 SQL Server 2012 Posted: 12 Jun 2013 04:53 AM PDT Hi,I am experiencing instances where not all the cores are being utilize for certain queries in SQL Server 2012. When this happens one of the cores becomes max to 100% CPU usage. This causes everything on the server to slow down and no one else can use the sql server until the offending query is kill from the server. My guess is that it is related to parallelism.Any suggestions on how to go about debugging this? I can't see the actual execution plan because we always end up killing the offending query. It is a production machine so I can't really play around with it. Thanks, |
| You are subscribed to email updates from SQLServerCentral / SQL Server 2012 / SQL 2012 - General To stop receiving these emails, you may unsubscribe now. | Email delivery powered by Google |
| Google Inc., 20 West Kinzie, Chicago IL USA 60610 | |
Weekly SQL Server newsletter with articles, forum posts, and blog posts via email. Subscribers receive our white paper with performance tips for developers.
sqlteam.com
