Yes, it will impose a major concern. Bandwidth/capacity and throughput. The LAN users will see excellent performance but it is possible and very likely that users across a WAN will experience poor performance due to either the capacity of their links or the throughput which may be due to high latency if the clients are a long distance from the server. You should have all of the processing done at the server tier(s) and only the results returned to the users if possible. The client software should be as light as possible and do as very little processing as necessary to generate the needed results.
Last Wiki Answer Submitted: August 18, 2009 2:10 pm by Labnuke9932,645 pts.
All Answer Wiki Contributors: Labnuke9932,645 pts.
If you live outside the United States, by submitting your email address you consent to having your personal data transferred to and processed in the United States.
Can you provide more details about what exactly you are looking for? Are there specific limitations you are concerned about, and if so, what specifically? Thanks!
More important (IMHO) that doing all the processing at the server, is to do all the processing where ever the data is. If the users enters a bunch of stuff, and the server only needs the resulting bit value, do the processing on the client side, and send the final bit to the server. If the server contains the bulk of the data, then do the work there sending just the small set of final results to the client.
Cache as much data as possible on the client side. If they will be downloading the same looking tables thousands of times a time, either cache that within the browser somewhere, checking to see if there are new values on the server as needed or setup a lookup database on the local site where the app can get the values from.
MrDenny – you raise a good point about process where the data is located. The only downside with that might be that the processing power may not be located there and that is the challenge in designing client/server and data networks. Caching is also a great performance enhancing technique as long as the cache data is fresh and relevant.
client /server architecture is kind of architecture where you put your data in database engine called server. the client requst the data from it e.g pc
Can you provide more details about what exactly you are looking for? Are there specific limitations you are concerned about, and if so, what specifically? Thanks!
More important (IMHO) that doing all the processing at the server, is to do all the processing where ever the data is. If the users enters a bunch of stuff, and the server only needs the resulting bit value, do the processing on the client side, and send the final bit to the server. If the server contains the bulk of the data, then do the work there sending just the small set of final results to the client.
Cache as much data as possible on the client side. If they will be downloading the same looking tables thousands of times a time, either cache that within the browser somewhere, checking to see if there are new values on the server as needed or setup a lookup database on the local site where the app can get the values from.
MrDenny – you raise a good point about process where the data is located. The only downside with that might be that the processing power may not be located there and that is the challenge in designing client/server and data networks. Caching is also a great performance enhancing technique as long as the cache data is fresh and relevant.
what is client server architecture
client /server architecture is kind of architecture where you put your data in database engine called server. the client requst the data from it e.g pc