Hi all, Thanks for all your responses. Maybe it'll be better if I described our use case in bit more in details. Our users send us trade file (sometimes has ~50K trades in it currently). And let say all of the trades have an invalid broker code, and all got kicked out, the user will have to go in and change the broker code 1 by 1 & that could take them hours. Therefore, the ability for our users to perform mass actions on the dataset (either client side or server side) would be a huge win for us.
Once again, thanks all. Ban --- In [email protected], "hank williams" <[EMAIL PROTECTED]> wrote: > > On Nov 12, 2007 4:35 AM, Tom Chiverton <[EMAIL PROTECTED]> wrote: > > > On Friday 09 Nov 2007, letterpigeon wrote: > > > Hi all, > > > requirement is for the grid to be able to support up to 100K+ or even > > > 1 million rows > > > > Why ? It's utterly impossible for a user to cope with this much data at > > once, > > so what's the point ? > > > > In general, the question of where to process data is totally legit. If you > read carefully, he did not say he wanted to *show* 100k rows at once. He > said he wanted to be able to do mass operations on it on the client side. > Without understanding his application and architecture I cannot opine on > whether it is better to do this on the client or server. But the idea, in > applications that may need to scale, of doing more work on the client, is a > good one. If you have lots of clients you can indeed save money on server > costs by having the clients help with the work. > > Hank >
