Using your web browser to obtain a Web page containing a small HTML file referring to...

 

Transcribed Image Text:

Using your web browser to obtain a Web page containing a small HTML file referring to 12 objects, each of size 100K bits. Suppose initially the IP address of the Web server is unknown (same initial condition for each of the following questions). Suppose that your host needs to visit all types of DNS servers (one per type) to retrieve the IP address. Each visit takes an RTT of 10 ms. The path between your host and the Web server has an RTT of 50 ms and a bottleneck link capacity of 1 Mbps. a) How long does it take to obtain the base HTML? (hint: ignore transmission time) b) How long does it take to obtain the entire Web page using non-persistent HTTP via a single TCP connection? c) Repeat b) for using 5 parallel connections. d) How long does it take to obtain the entire Web page using persistent HTTP without pipelining? e) Repeat d) for persistent HTTP with pipelining. f) What is the minimum number of parallel connections that non-persistent HTTP needs to be faster than persistent HTTP without pipelining? Using your web browser to obtain a Web page containing a small HTML file referring to 12 objects, each of size 100K bits. Suppose initially the IP address of the Web server is unknown (same initial condition for each of the following questions). Suppose that your host needs to visit all types of DNS servers (one per type) to retrieve the IP address. Each visit takes an RTT of 10 ms. The path between your host and the Web server has an RTT of 50 ms and a bottleneck link capacity of 1 Mbps. a) How long does it take to obtain the base HTML? (hint: ignore transmission time) b) How long does it take to obtain the entire Web page using non-persistent HTTP via a single TCP connection? c) Repeat b) for using 5 parallel connections. d) How long does it take to obtain the entire Web page using persistent HTTP without pipelining? e) Repeat d) for persistent HTTP with pipelining. f) What is the minimum number of parallel connections that non-persistent HTTP needs to be faster than persistent HTTP without pipelining?