CODEWITHSUNDEEP
c#performancedatatable
Shahin
Shahin

Reputation: 12851

Fastest way to update datatable

Assume we have a DataTable with 100 columns and 100 rows. We have a dictionary of values like this that contains 100 members: 

Dictionary<int, string> dictionary = GetValues();

In above dictionary key is index of the Datatable and value of dictionary is target value that we want to update in Datatable.

What's the fastest way to update the Datatable? I use this way but it's slow : 

      foreach (var pair    in dictionary)
        {

            timeTable.Rows[pair.Key].SetField(columnIndex, pair.Value);
        }

Assuming we have to update all rows of 40 columns, it doesn't have good performance. I was thinking that Datatable is like a Matrix and it shouldn't take time at all when you want to update a cell on Matrix while you have indexes of the row and the column.

Upvotes: 0

Views: 2731

Answers (1)

quantdev
quantdev

Reputation: 23813

DataTable is a very expensive structure that dos a lot behind the scenes (particularly because they are designed to handle anything).

If performance is critical consider:

  • Creating your own structure, such as a class encapsulating an array of strings arrays (e.g. a strings matrix, on which access will be very fast)

  • Udating your table in parallel (since you are never accessing the same row), knowing that DataTable is NOT thread safe for writing operations.

e.g. 

// Probably not thread safe, after the MSDN documentation
Parallel.ForEach(Dictionary, pair =>
    {
      timeTable.Rows[pair.Key].SetField(columnIndex, pair.Value);
    });

EDIT: This will not work in the general case (write operations even on distinct rows might affect global state of the DataTable, so I suggest that you go with your own structure if performance is critical, and if you don't need specific features of the DataTable)

Upvotes: 1

Related Questions