What sort of bench marks have any of you recieved when working with large datasets.
I am working with groundwater levels with datasets upwards of 30,000 points over time. I have yet to render a Chart from a List<> without incurring an OutOfMemoryException on a dual core with 4GB of RAM.
Per the design of RadChart, what size datasets were in mind when creating the control, and how many points are considered "feasible".
Many thanks,
--
I am working with groundwater levels with datasets upwards of 30,000 points over time. I have yet to render a Chart from a List<> without incurring an OutOfMemoryException on a dual core with 4GB of RAM.
Per the design of RadChart, what size datasets were in mind when creating the control, and how many points are considered "feasible".
Many thanks,
--
3 Answers, 1 is accepted
0
Accepted
Hello Ryan,
In order to get a correct chart image you will need at least 1 pixel for each datapoint, so we have not considered adding more than a thousand or two thousand datapoints to a chart due to the increasing image size. RadChart will render 5000 datapoints within half a minute, but using such or bigger amount of datapoints seems like overkill. A workaround would be to combine several data points in one chart series item using the average value.
We understand that the work-around offered requires substantial effort on your part, however, RadChart was primarily designed with ease-of-use in mind and currently cannot scale this much. We realise this shortcoming and have included plans in our long-term charting strategy to address such situations, but currently the presented work-around is the only suggestion we can offer.
Best Regards,
Ves
the Telerik team
Instantly find answers to your questions at the new Telerik Support Center
In order to get a correct chart image you will need at least 1 pixel for each datapoint, so we have not considered adding more than a thousand or two thousand datapoints to a chart due to the increasing image size. RadChart will render 5000 datapoints within half a minute, but using such or bigger amount of datapoints seems like overkill. A workaround would be to combine several data points in one chart series item using the average value.
We understand that the work-around offered requires substantial effort on your part, however, RadChart was primarily designed with ease-of-use in mind and currently cannot scale this much. We realise this shortcoming and have included plans in our long-term charting strategy to address such situations, but currently the presented work-around is the only suggestion we can offer.
Best Regards,
Ves
the Telerik team
Instantly find answers to your questions at the new Telerik Support Center
0
Steve
Top achievements
Rank 1
Rank 1
answered on 17 Jul 2008, 05:55 AM
Thanks for the prompt response.
Unfortunately, excluding data points is not an option due to the nature of the data and purpose of the chart implementation.
Rolling my own chart I can get 50,000 data points in approximately 220ms.
Of course, my version has about 1/1000 of the features that the Telerik Chart, so I can see why it is able to accomplish these numbers... Really like the zooming capabilities, keep up the great work!
I appreciate your time, many thanks,
RA
Unfortunately, excluding data points is not an option due to the nature of the data and purpose of the chart implementation.
Rolling my own chart I can get 50,000 data points in approximately 220ms.
Of course, my version has about 1/1000 of the features that the Telerik Chart, so I can see why it is able to accomplish these numbers... Really like the zooming capabilities, keep up the great work!
I appreciate your time, many thanks,
RA
0
jc mag
Top achievements
Rank 1
Rank 1
answered on 07 Aug 2008, 11:56 AM
I need to display 65536 points on a plot chart ; it doesn't scale with RadChart. Do you know other components that could be used for my need?