This question is locked. New answers and comments are not allowed.
Hi,
We have our own highly optimized data set implementation with an API somewhat similar to .NET's DataTable. We have integrated it with the RadGridView using the approach suggested by Vlad . While it seems to work, the aggregation / filtering / searching operations are way too slow even on small data sets (<10K rows).
I understand that you can't do much since extracting a value from the data set is somewhat expensive (virtual methods, boxing/unboxing, the need to store it, etc), but our data source is specifically optimized for doing such things with blazing speed and no additional storage.
The question is, is there some API that we could utilize to make RadGridView utilize our optimized routines?
Regards,
Andrew
We have our own highly optimized data set implementation with an API somewhat similar to .NET's DataTable. We have integrated it with the RadGridView using the approach suggested by Vlad . While it seems to work, the aggregation / filtering / searching operations are way too slow even on small data sets (<10K rows).
I understand that you can't do much since extracting a value from the data set is somewhat expensive (virtual methods, boxing/unboxing, the need to store it, etc), but our data source is specifically optimized for doing such things with blazing speed and no additional storage.
The question is, is there some API that we could utilize to make RadGridView utilize our optimized routines?
Regards,
Andrew
2 Answers, 1 is accepted
0
Hello Andrew,
Vlad
the Telerik team
Generally if the source of the grid is IEnumerable and all properties (and their types) can be extracted normally everything should work as expected - grouping, filtering, sorting, aggregates, etc.
Can you post more info about your data layer?
Vlad
the Telerik team
Explore the entire Telerik portfolio by downloading the Ultimate Collection trial package. Get it now >>
0
Andrew
Top achievements
Rank 1
Rank 1
answered on 02 Nov 2011, 05:19 PM
Hi Vlad,
As I said, it works; it is performance we're concerned with. On millions of rows, Telerik's aggregation and filtering takes seconds, while the corresponding methods of our data storage work instantaneously; if only we could combine rich features of your grid with the performance of our data storage, it'll be a huge win! Regarding the data layer - it is a strongly-typed, column-based in-memory storage optimized for aggregations and filtering.
Cheers,
Andrew
As I said, it works; it is performance we're concerned with. On millions of rows, Telerik's aggregation and filtering takes seconds, while the corresponding methods of our data storage work instantaneously; if only we could combine rich features of your grid with the performance of our data storage, it'll be a huge win! Regarding the data layer - it is a strongly-typed, column-based in-memory storage optimized for aggregations and filtering.
Cheers,
Andrew