Multitasking in C Sharp

Aniket Narvankar

I am updating large number of records one by one using c sharp. The requirement is such that I have to write sql query in c sharp. Requirement compare records from table 1 with records in table 2 on the basis of Id,if found in table 2 update in table1,not found set message Id does not exists,if got exception update exception message and this I am doing one by one by using foreach loop to compare. But In case of large number of records it takes time for 100000 records it takes 2 Hours,how to use multitasking to make processing faster.

4 Replies

Know the answer? Post it — somebody with the same question will find it here.

Priyanka K S5y ago

Hi,

First, you can get all the missing messageId from the table1, it's better if you use Stored Procedure for this.

SELECT message.ID
FROM table1
WHERE message.ID NOT IN(SELECT message.ID FROM table2)

Consider there might be 200 missing MessageId.

Create a function Generic function [GenerinFun()] to do update table operation.

In from Main() Function create multiple tasks for every 10 or 20 messageId using LINQ pagination :


//used by LINQ to SQL  
   public static IQueryable Page(this IQueryable source, int page, int pageSize)  
   {  
       return source.Skip((page - 1) * pageSize).Take(pageSize);  
   }  
Now you have multiple task , run all tasks concurrently.
await Task.WhenAll(tasks);  
I just try give you rough picture how we can solve this problem.
Hope this helps!!

Srinivasan Ramamoorthi5y ago
Can you try bulk update operation in C#? That should fit your requirement
+1
Satish Vishwakarma5y ago
use user defined table
+1
Mahesh Chand5y ago
It should not take you hours unless somewhere your code is slow, like update table or sending a message. Update Table and send message can be put in a secondary thread if they are slow.

Ideally, you should not compare one by one and update one by one. Get both table in memory (all records), say DataTables and compare them. Update Datatable in memory and send data back when all updates are done. It will be one time update.

Most cases, SQL queries are slow. Check Query performance and table indexing for performance.

Cheers!
+1