We have data which changes every week. We had considered seperating
historical records and current records into two different tables with
the same columns, but thought it might be simpler to have them all
together in one table and just add a WeekID int column to indicate
which week it represents (and perhaps an isCurrent bit column to make
querying easier). We have a number of tables like this, holding weekly
data, and we'll have to query for historical data often, but only back
through the last year -- we have historical data going back to 1998 or
so which we'll rarely if ever look at.
Is the all-in-one-table approach better or the seperation of current
and historical data? Will there be a performance hit to organizing data
this way? I don't think the extra columns will make querying too much
more awkward, but is there anything I'm overlooking in this?
I need some advise on how to create a historical database.
What is the best way of doing this? For example, should I create a new row if a column is changed in a row and Time Stamp all record? What happens when I have child tables link to a Header table?
I have been looking on the NET for methods of creating a historical database, but I cant find any.
From what I've read this is called 'slowly changing dimensions'. Bassically the system I'm working on needs to store the history of certain data so that at any time a user can look up an old project and view it exactly as is, even though the associated parts might have had certain changes over time. From what I can tell Type 2 ( current and historical records are stored in the same table) seems to be the most popular. Type 4 (current records in one table and historical records in a seperate history table) seems like it would also work but I've been unable to find any articles comparing the two. Does anybody have any info on the dis/advantages of one v.s. the other?
We have a database that adds over 100,000 records per day. This goes back to 2002 and I only need historical data for 6 months. Presently we can only can delete 1000 row at a time. Is there a faster way of deleting. We seem to continuely run out of disk space. Urgent!!!!!!!!!!!
I have a SQL2005 db for tracking the prices of products at multiple retailers. The basic structure is, 'products' table lists individual products, 'retailer_products' table lists current prices of the products at multiple retailers, and 'price_history' table records when the price of a product changes at any retailer. The prices are checked from each retailer daily, but a row is added to the 'price_history' only when the price at the retailer changes.
I have the following query to retrieve the price history of a given product at multiple retailers:
SELECT price_history.datetimeofchange, retailer.name, price_history.price FROM product, retailer, retailer_product, price_history WHERE product.id = 'b486ed47-4de4-417d-b77b-89819bc728cd' AND retailer_product.retailerid = retailer.id AND retailer_product.associatedproductid = product.id AND price_history.retailer_productid = retailer_product.id
This gives the following results:
2008-03-08 Example Retailer 22.3 2008-03-28 Example Retailer 11.8 2008-03-30 Example Retailer 22.1 2008-04-01 Example Retailer 11.43 2008-04-03 Example Retailer 11.4
The question(s) I have are how can I:
1 - Get the price of a product at a given retailer at a given date/time For example, get the price of the product at Retailer 2 on 03/28/2008. Table only contains data for Retailer 1 for this date, the behaviour I want is when there is no data available for the query to find the last data at which there was data from that retailer, and use the price from that point - i.e. so for this example the query should result in 2.3 as the price, given that was the last recorded price change from that retailer (03/08/2008).
2 - Get the average price of a product at a given retailer at a given date/time In this case we would need to perform (1) across all retailers, then average the results
I've got a customer who wants reproducible/historical reporting. The problem is that the underlying data changes.
I tried to explain that this can't be done (can it?), but he doesn't understand.
To illustrate the situation - Let's say a teacher wants to track spelling test scores for her students. The below are scores for students A, B, and C (for January, February, March)
A: {70,80,85} B: {70,65, 80} C: {100,90,100}
So, I can generate a historical report that charts the class average and student trend - that's pretty easy.
Now, in April, we find that the school board has mandated that the British spelling of words is ok, so now the cumulative scores (for January, February, March, April)
He wants a report showing the January average as (70+70+100)/3 = 80, when really it is (90+80+100)/3 = 90.
Now imagine that there are actually thousands of data points changing like this... Now also imagine that we add and remove students on a regular basis...
He and his office manager get frustrated when I explain that the reports are not simple - in their mind it is. They have determined the solution is to get a report writer and buy Crystal Reports... I've tried to explain that the problem is that the report specification is unclear (basically - they don't understand what they want). The situation is ok for now, I'm just trying to plan for when they figure out that buying Crystal Reports won't change their situation (except they are done several thousand dollars)...
I have a database which contains time series data (historical stock prices) which I have to search for patterns on a day to day basis. But searching this historical data for patterns is very time consuming not only in writing the complex t-sql scripts but also executing them.
Table structure for one min data: [Date] [Time] [Open] [High], [Low], [Close], [Adjusted_Close], [MA], [DI]..... Tick Data: [Date] [Time] [Trade] Most time consuming queries are with lots of inner joins. So for example if I have to compare first few mins data then I have to do inner join like: With IntervalData AS ( SELECT [Date], Sum(CASE WHEN 1430 = [Time] THEN [PriceRange] END) AS '1430', Sum(CASE WHEN 1431 = [Time] THEN [PriceRange] END) AS '1431', Sum(CASE WHEN 1432 = [Time] THEN [PriceRange] END) AS '1432' FROM [INDU_1] GROUP BY [Date] ) SELECT [Date] ,[1430], [1431], [1432], [1431] - [1430] As 'Range' from IntervalData WHERE ([1430] > 0 AND [1431] < 0 AND [1432] < 0) OR ([1430] < 0 AND [1431] > 0 AND [1430] > 0) ------------------------------------------------------------------------ select ind1.[Time], ind1.PriceRange,ind2.[Time], ind2.PriceRange from INDU_1 ind1 INNER JOIN INDU_1 ind2 ON ind1.[Time] = ind2.[Time] - 1 AND ind1.[Date] = ind2.[Date] where (ind1.[Time] = 2058) AND ((ind1.PriceRange > 0 AND ind2.PriceRange >0) OR (ind2.PriceRange < 0 AND ind1.PriceRange < 0)) ORDER BY ind1.[Date] DESC; Is there anyway I can use Sql 2005 Data mining models to make this searching faster?
I am working on a project, which involves displaying trends of certain aggregate values over time. For example, suppose we want to display how the number of active and inactive users changed over time.
One issue is how to store historical data. First of all, should I create a separate database for each historical snapshot or should I use one database for all snapshots? Second, our database size is a couple of gigabytes and replicating the entire database on a daily basis is not feasible. An alternative solution is to back up aggregate values, but how do I back up results of aggregate queries, where the user can specify a date range in the WHERE-clause? Another solution is to create fact tables from our relational schema and back those up.
Another issue is how to query historical data. Using multiple databases to store historical snapshots makes it harder to query.
As you can see there are several design alternatives and I would like to know how this sort of problem is generally solved in the industry. Does SQL Server provide any support for solving this problem?
My application is to capture employee locations.Whenever an employee arrives at a location (whether it is arriving forwork, or at one of the company's other sites) they scan the barcode ontheir employee badge. This writes a record to the tblTSCollected table(DDL and dummy data below).The application needs to be able to display to staff in a control roomthe CURRENT location of each employee.[color=blue]>From the data I've provided, this would be:[/color]EMPLOYEE ID LOCATION CODE963 VB002964 VB003966 VB003968 VB004977 VB001982 VB001Note that, for example, Employee 963 had formerly been at VB001 but wasmore recently logged in at VB002, so therefore the application is notconcerned with the earlier record.What would also be particularly useful would be the NUMBER of staff ateach location - viz.LOCATION CODE NUM STAFFVB001 2VB002 1VB003 2VB004 1Can anyone help?Many thanks in advanceEdwardNOTES ON DDL:THE BARCODE IS CAPTURED BECAUSE THE COMPANY MAY RE-USE BARCODE NUMBERS(WHICH IS DERIVED FROM THE EMPLOYEE PIN), SO THEREFORE THE BARCODECANNOT BE RELIED UPON TO BE UNIQUE.THE COLUMN fldRuleAppliedID IS NULL BECAUSE THAT PARTICULAR ROW HAS NOTBEEN PROCESSED. THERE ARE BUSINESS RULES CONCERNING EMPLOYEE HOURSWHICH OPERATE ON THIS DATA. ONCE A ROW HAS BEEN PROCESSED FORUPLOADING TO THE PAYROLL APPLICATION, THE fldRuleAppliedID COLUMN WILLCONTAIN A VALUE. IN THE PRODUCTION SYSTEM, THEREFORE, ANY SQL ASREQUESTED ABOVE WILL CONTAIN IN ITS WHERE CLAUSE (fldRuleAppliedID IsNULL)if exists (select * from dbo.sysobjects where id =object_id(N'[dbo].[tblTSCollected]') and OBJECTPROPERTY(id,N'IsUserTable') = 1)drop table [dbo].[tblTSCollected]GOCREATE TABLE [dbo].[tblTSCollected] ([fldCollectedID] [int] IDENTITY (1, 1) NOT NULL ,[fldEmployeeID] [int] NULL ,[fldLocationCode] [varchar] (50) COLLATE SQL_Latin1_General_CP1_CI_ASNULL ,[fldTimeStamp] [datetime] NULL ,[fldRuleAppliedID] [int] NULL ,[fldBarCode] [varchar] (50) COLLATE SQL_Latin1_General_CP1_CI_AS NULL) ON [PRIMARY]GOINSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (963, 'VB001', '2005-10-18 11:59:27.383', 45480)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (963, 'VB002', '2005-10-18 12:06:17.833', 45480)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (964, 'VB001', '2005-10-18 12:56:20.690', 45481)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (964, 'VB002', '2005-10-18 15:30:35.117', 45481)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (964, 'VB003', '2005-10-18 16:05:05.880', 45481)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (966, 'VB001', '2005-10-18 11:52:28.307', 97678)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (966, 'VB002', '2005-10-18 13:59:34.807', 97678)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (966, 'VB001', '2005-10-18 14:04:55.820', 97678)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (966, 'VB003', '2005-10-18 16:10:01.943', 97678)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (968, 'VB001', '2005-10-18 11:59:34.307', 98374)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (968, 'VB002', '2005-10-18 12:04:56.037', 98374)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (968, 'VB004', '2005-10-18 12:10:02.723', 98374)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (977, 'VB001', '2005-10-18 12:05:06.630', 96879)INSERT INTO dbo.tblTSCollected(fldEmployeeID,fldLocationCode,fldTimeStamp,fldBarCode)VALUES (982, 'VB001', '2005-10-18 12:06:13.787', 96697)
I want to load historical data from an old system into a new one.Thing is, that old system stored dates as Datetime and the new one uses DateTimeOffset.
All data was collected in the same Time Zone... but with the Daylight Saving Time (DST)
The offset is either +04:00 or +05:00, based on the calendar date. To add to the complexity, the rules for DST changed a couple of years ago.
To determine the offset, I'd need to know what was or would have been the server Timezone for each historical date.
Recently, I partitioned one of my largest tables into multiple monthly field groups. For the current month, it is attached to my "Active' table. The older records are kept in the "historical" table. I need an efficient way to pull records when have a date range that can be spread across both tables.
I'm looking to get counts on historical data where the number of records exists on or after May 1 in any given year. I've got the total number of records for each year worked out, but now looking for the number of records exist after a specific date. Here's what I have so far.
SELECT p.FY10,p.FY11,p.FY12,p.FY13,p.FY14,p.FY15 FROM ( SELECT COUNT(recordID) AS S, CASE DateFY
I am getting ready to start a project where I am charged with moving out old data from production into a newly created historical DB. We have about 8 tables that are internal audit tables, that are big and full of old data. These tables are barely used and are taking up way too much space and time for maintenance.
I would like to create a way (SSIS?) to look at the date field in each of the 8 tables and copy out anything older than two years into my newly created history DB. Then deleting the older records from the source DB.
I don't know if SSIS is the best method to use. If it is, what containers to use to move over data, then how to do delete from source?
Can I do the mass deletes on my audit source tables without impacting performance/indexes/fragmentation?
Ok, I'm looking to get counts on historical data where the number of records exists between two dates with different years. The trick is the that the dates fall in different years. Ex: Give me the number of records that are dated between 0ct 1, 2013 and July 1, 2014.
A previous post of mine was similar where I needed to get records after a specific date. The solution provided for that one was the following. This let me get any records that occured after May 1 per given Fiscal year.
SELECT MAX(CASE WHEN DateFY = 2010 THEN Yr_Count ELSE 0 END) AS [FY10], MAX(CASE WHEN DateFY = 2010 THEN May_Count ELSE 0 END) AS [May+10], MAX(CASE WHEN DateFY = 2011 THEN Yr_Count ELSE 0 END) AS [FY11], MAX(CASE WHEN DateFY = 2011 THEN May_Count ELSE 0 END) AS [May+11], MAX(CASE WHEN DateFY = 2012 THEN Yr_Count ELSE 0 END) AS [FY12],
[Code] ....
I basically need to have CASE WHEN MONTH(OccuranceDate) between Oct 1 (beginning year) and July 1 (ending year).
I have a question about historical table. I have a table in Sql Server that keeps the history of my clients for the last 3 months. This table has a PK : client_code and rep_date_id. I want to do a query that returns me a client and to extract the date when i inregistreted for the first time.
I think about a cursor, but i don't know how to use it. My table looks like this structure: - client_code -activity_code -country_code -district_code -... -rep_date_id_n
And I want to return --activity_code -country_code -district_code -... -start_date -current_date
How to build a TRIGGER that copies, all the time, the data from the table onwhich the transaction occurs to the historical table?thanksFernand---Outgoing mail is certified Virus Free.Checked by AVG anti-virus system (http://www.grisoft.com).Version: 6.0.659 / Virus Database: 423 - Release Date: 2004-04-15
I have about 45000 records in a CSV file, which I am using as HTTP request parameters to query a website and store some results in a database. This is the kind of application which runs 24/7, so database grows really quickly. Every insert fires up a trigger, which has to look for some old records based on some criteria and modify the last inserted record. My client is crazy about performance on this one and suggested to move the old records into another table, which has exactly the same structure, but would serve as a historical table only (used to generate reports, statistics, etc.), whilst the original table would store only the latest rows (so no more than 45k at a given time, whereas the historical table may grow to millions of records). Is this a good idea? Having the performance in mind and the fact that there's that trigger - it has to run as quickly as possible - I might second that idea. Is it good or bad? What do you think?
I read a similar post here, which mentioned SQL Server 2005 partitioning, I might as well try this, although I never used it before.
I have an SCD that contains a historical output path. If I throw dataviewer in the flow before the SCD, I can see that it should trigger a trip down that lane, but its not. In the advanced editor, all looks good. Anything I can check. BTW, the datatype of the fields that should cause the SCD are datetime.
I have DWH where users query on regular basis. I wanted to know what queries they run and which user fired the queries for the last one month or 6 months.
I have a simple table with three fields: ID, LastName, FirstName. The ID is defined as the PK. In the table is a record of "12345, Smith, John". The incoming flat file has a record of "12345, Smith, Johnny".
In the SCD transform, the ID is the business key, and Last Name and First Name are defined as historical attributes.
During the load, the SCD transform correctly sends the data down the right path, but the insert fails with a primary key violation - as I would expect since it's trying to create a new current record.
How do I get around this problem without removing the PK ???
We have an issue in a SCD where a number of records may be presented that have changes to their attributes of EVERY type.
Example,
BusinessKey: xxxxxxxx
BuildingTypeId: 7
BusinessUnitHistoryId: 4019
BusinessUnitId: 4019
CurrencyId: 26
DevelopmentTypeId: 14
MarketId: 182
Name: abcdefgh
CurrencyId is a fixed attribute MarketId & BuildingTypeId and the BusinessUnitId & BusinessUnitHistoryIds are historical attributes Name is a changing attribute
The behaviour of the ETL seems to suggest that if fixed attribute changes are detected, these rows will error and therefore the changing & historical attributes will NOT be amended during the SCD transformation. Is this correct... as it seems to be what is happening.
Is it possible to design a Cube with a Dimension "DimEmployee" (SCD Type2) and Query the following:Â What "title" had the person "xyz" at time "2015-01-01"? - See Example data on Images.
I have also a DimTime with Day granularity. The DimEmployee is not at day granularity.
I'm trying to create a website based on interesting historical facts, Searchable by date. I need each fact entry to have a datetime field in the database, but unfortunately this method only lets me go as far back as the year 1753 (for an inexplicable reason, I can go up to the year 9999).Is there something I'm overlooking here, or did Microsoft goof this one? Will I have to store my dates as 3 separate custom month/day/year fields? If anybody has a solution for this, please let me know (a B.C./A.D enabled solution isn't at all needed, but I'm open to that too).
I have a historical table on a dedicated SQL Server (let's call it the reporting db) that is populated every morning with production data that does not already exist. The data in the prod table is purged after 7 days and nothing is ever deleted from the historical table. I have set up the linked server between the two 2008 SQL Servers, but when I try to run this simple query from the reporting DB, it takes more than 5 minutes and still "executing". I eventually have to cancel it:
-- INSERT INTO Temp_Import_historical SELECT TOP 1 * FROM [192.168.1.100].ProdDB.dbo.Temp_Import_historical a WHERE NOT EXISTS (select [Temp_Import_ID] from Temp_Import_historical where a.[Temp_Import_ID] = Temp_Import_historical.[Temp_Import_ID])
I have omitted the INSERT statement on purpose, since I can't even get to output 1 row. Why this is such a resource intensive query?
I have built a time series model to forecast sales value
I have data from jan 2004 to jan 2006 and the sales value is at a day level in my database. But I am aggregating it to month level in the DSV of the mining model.
I am required to make only historical predictions using the above model starting form jan 2004 to jan 2006 for every month.
I have set Historical_Model_Count and Historical_Model_Gap parameter values to 24 and 10 respectively, and trying to predict for the past few months (PredictTImeseries(SalesValue,-1,1))
But its throwing me the following error
Error(Data Mining): A time series prediction was requested with a start time further in the past than the internal models of the mining model, Sales Forecast, specified in the HISTORIC_MODEL_GAP and HISTORIC_MODEL_COUNT parameters can process
In fact it throws the above error irrespective of what the Historical_Model_Count and Historical_Model_Gap parameter values are
I am not able to figure our why this problem is happening?
What should the parameter values for the above scenario?
It would also be helpful if I can get an explanation on how these two parameters affect the historical predictions. I kind of understand that these two parameters are important for historical predictions but don€™t know why or how.
I am new to SSAS and i want to try to build a "Sales" model. I will have some "Usage" data for some timespanns, but I am not quite sure how to tackle this. Is there somewhere a "Howto" for this?
Edit: There are several locations, and for each location a forecast is needed. And the Icing would be If I would be able to tell where my supplies must go 1st to achieve the best sales...
The potential Client wants to use Oracle but I would like to show them that SQL Server is the better tool for this ;)
I have a Slowly changing dimension that I am using to populate a dimension table. My problem is this when I run the package and any of the fields are marked as Historical Attributes it will add an additional row regardless of the fact that the incoming data and the data in the warehouse match exactly.
I've tried several things to fix this problem but so far none of them have worked. Some of the things I have tried that haven€™t worked are to match all the data types (which I have to do anyways) I've tried trimming the strings, I've also tried adding just one column
I am using a data conversion to convert them from varchar (the source datatype) to nvarchar(the warehouse datatype)
I'm at a dead end here and don't know where to go any help would be greatly appreciated.
I have 2 identical tables one contains current settings, the other contains all historical settings.I could create a union view to display the current values from table A and all historical values from table B, butthat would also require a Variable to hold the tblid for both select statements.
Q. Can this be done with one joined or conditional select statement?
DECLARE @tblid int = 501 SELECT 1,2,3,4,'CurrentSetting' FROM TableA ta WHERE tblid = @tblid UNION SELECT 1,2,3,4,'PreviosSetting' FROM Tableb tb WHERE tblid = @tblid
I am wondering if it is possible to use SSIS to sample data set to training set and test set directly to my data mining models without saving them somewhere as occupying too much space? Really need guidance for that.
I have used both data readers and data adapters(with datasets) in the projects that I have worked on. I am trying to get some clarification on when I should be using which one. I think I am doing this correctly but I want to be sure I am developing good habits.
As the name might suggest, it seems like a datareader is for only reading data. I have read that the data adapter and dataset are for a disconnected architecture. Or, that they can be used for this type of set up. I have been using the data adapter and datasets when writing to a database and the datareader when reading from a database.
Is this how these should be used? Is the data reader the best choice for reading data? Am I doing this the optimal way from a performance stand point?
......................................................thanks in advance
We already integrated different client data to MDS with MS Excel plugin, now we want to push back updated or new added record to source database. is it possible do using MDS? Do we have any background sync process to which automatically sync data to and from subscriber and MDS?
