I've been fighting with a solution - I have to take a flat file, sometimes 200 mb or larger and and display that in a grid.
Stripping the data from the file is a snap using any one of several methods, but getting into a useful object the Grid can use becomes a real time-crusher.
The raw data is comma-delimited text with headers.
Is it possible to feed a csv into the grid? If so, how?
I can't provide sample code because of security and time constraints.
The most promising approach has been query the the target file using OleDb classes including a OleDbDataReader. Reading a 400k records file into a data table takes about 20 seconds give or take. Applying that to a Grid just blows everything out of the water. Virtual scrolling doesn't help. I simply can't get a large file to display in your grid helper.
Grid for JS doesn't work any better.
Can the Grid actually display a datasource of that size in a reasonable amount of time? Like less than a minute?
1 Answer, 1 is accepted
Passing 400K data items to the client to be handled by the Grid DataSource will result in slow performance and possibly a browser crash in IE. This is too large data to keep in-memory in a browser. Theoretically, 400K data items could be passed to the Grid, like in this example https://dojo.telerik.com/iRODafof, but the result depends on many things:
- the size of the data items themselves (number and size of fields)
- the client-side features enabled in the Grid
- templates in the Grid columns
To get reasonable behavior with large data like that, data should be kept on the server in a temporary store or copied to a database to be requested by the Grid using server-side paging (returning only the current page of data on the client) or virtual scrolling with server-side paging of the DataSource. In this way, only small portions of the data will be returned and loaded on the client during paging.
Regards,
Tsvetina
Progress Telerik