Let me start with a simple question. Suppose you want to have a bicycle. How do you want it? Getting the bicycle in one piece from one shop or to have each part from different shop and get it assemble by some mechanic. Yah you got the point. Obviously, you want the bicycle in one piece and readymade.
Employment is the biggest buzzword in the market and yet is the most precious dream of every youngster. The young generation coming out of colleges / institutes with lots of ener
gy and enthusiasm to grow up in the industry and to glow up their future in their respective fields and interests, perhaps don’t know about the up and down and the working mechanism of all this. Most of the people are just concern about their size of the pocket at the end of the month, especially in the IT world.
In my previous posts << SSIS vs Text File Importing –I >> and << SSIS vs Text File Importing –II >> , we discussed about the Row Delimiter issue of SSIS while dealing with text files and the workaround for the same. In this post I will show you how to achieve the same through SSIS package.
In my previous post named “<<SSIS vs Text File Importing –I>>”, we discussed about the problem statement for the strange behavior of SSIS over importing of Text Files. In this post, I will disclose the reason behind it. Let’s see how Excel and SSIS treats text files while importing data from it.
Last week I come across a serious problem with SSIS while dealing with text files especially with tab delimited text files. It is showing the wrong data after importing into SQL SERVER database. I was shocked by this behavior of SSIS. The total row count in the destination table is much smaller than the source file row count. When I opened the same file in Excel or notepad, I saw the data is correct and properly aligned. Then I deeply dived into the problem and tried to figure out what may be the cause for such an unexpected behavior. Even I am surprised more when I implemented the same process in DTS package (SQL 2000); there it is showing me correct number of records at the destination. Pretty strange, all three products i.e. SSIS, DTS and Excel are from the same manufacturer (Microsoft), but showing different behavior for same type of process (importing data from text file).
In SQL server 2000, scheduling a DTS package is not as easy as compare to SSIS packages in SQL 2005 or 2008. As the DTS packages are basically stored on the server itself and are being managed on the server only, it’s little tricky to schedule more than one packages in a single Job.
In a job we don’t have any options to directly link a DTS package other then “Operating System Command (CmdExec)” under which DTSRun command is used to run the package. The drawback here is the way Steps are linked to packages.
Here comes one more challenging yet interesting topic to tide over. The requirement goes something like this: You have some sources, let it be some sales data or some Call center data coming from different sources. Data can be of different media. But you are able to load it in your staging tables i.e. one staging database you are maintaining like StageDB for storing these incremental data.
So, here we have both master i.e. Dimensional data and Detail i.e. Fact data in our stage database. But before processing further and loading it into our Mart or Data warehouse, we need to check if there is any redundancy at the row level for each of these tables in the staging database. Here comes the main problem, what if you don’t know how many tables are there in the staging database and how many columns are there for each table in the staging database but still you have to keep only unique records for each table.
Problem: I have a table for customer information with Customer Ids and Customer Type. The customer may fall into different types i.e. the customer may belong to Type A or Type B or may belong to both the groups. The requirement is to find out the count of the customers as per the type in each group and if the customer falls under more than one group, it must not be counted under the individual group rather it should be counted under the group with both the types.
Problem:
The requirement is to get top 10 employee’s name randomly from each department in a company. Let me elaborate it. Suppose in a company ABC, there are 1500 employees with 5 departments and each department have more than 200 employees. The manager wants to list down at least 10 names of the employees randomly from each department for a survey.
Solution:
Don’t have an account? Create one, its simple
Recent Comments