Client Software Design Algorithms and Issues Chuan-Ming liu Computer Science and Information Engineering Spring 2004, NTUt TAIWAN CSIE NTUT Taiwan
CSIE, NTUT, Taiwan 1 Client Software Design Algorithms and Issues Chuan-Ming Liu Computer Science and Information Engineering Spring 2004, NTUT TAIWAN
Client Software Design a How applications become clients by initiating communication a How applications use TCP or UDP protocols to contact a server How they use socket calls to interact with the protocols CSIE NTUT Taiwan
CSIE, NTUT, Taiwan 2 Client Software Design ◼ How applications become clients by initiating communication ◼ How applications use TCP or UDP protocols to contact a server ◼ How they use socket calls to interact with the protocols
Learning algorithms Instead of Details Learn know-what instead of know -how first a After knowing what to do, finding out how to do is straightforward CSIE NTUT Taiwan
CSIE, NTUT, Taiwan 3 Learning Algorithms Instead of Details ◼ Learn know-what instead of know-how first ◼ After knowing what to do, finding out how to do is straightforward
Client Architecture a Client applications are simpler than server applications No concurrent interaction Most client software executes as a conventional application a Dont need to enforce protection CSIE NTUT Taiwan
CSIE, NTUT, Taiwan 4 Client Architecture ◼ Client applications are simpler than server applications ◼ No concurrent interaction ◼ Most client software executes as a conventional application ◼ Don’t need to enforce protection
Identifying servers location Have the server 's domain name or ip address as constants when program is compiled a Using address makes client software faster and less dependent on local computing environment no flexibility Using a generic name for the server and an alias to the domain name system for that name Improvement Obtain information from stable storage Can not execute program unless the file is available CSIE NTUT Taiwan 5
CSIE, NTUT, Taiwan 5 Identifying Server’s location ◼ Have the server’s domain name or IP address as constants when program is compiled ◼ Using address makes client software faster and less dependent on local computing environment – no flexibility ◼ Using a generic name for the server and an alias to the domain name system for that name – improvement ◼ Obtain information from stable storage ◼ Can not execute program unless the file is available