Suppose Alices web browser needs to download a web page of 5 Kbytes (HTML file size). Consider the network throughput between Alice and the web server to be 5 Mbps (for both upload and download), and propagation delay of 0.1 seconds within Alices ISP, and 0.5 seconds between Alices host and any host outside her ISP. Assume instantaneous processing at the server and client and negligible transmission delay for TCP and HTTP messages if without entity body.

a) Suppose that Alices web browser does not know the IP address of the web server. Please, write down the resource record (ignore the ttl field) for www.pieroweb.it, which has address 192.115.70.8, that is stored in three authoritative DNS servers, dns1.pieroweb.it, dns2.pieroweb.it, dns3.pieroweb.it, which have addresses 192.115.70.8, 192.115.70.9, 192.115.70.10, respectively. Write down the resource records that would be contained in the TLD it servers to allow the translation of the web server address www.pieroweb.it (suppose no DNS caching) when contacted by Alices web browser.

c) Suppose that the web page has references to 5 other web objects of 5Mbytes each to be visualized in the page. How long does it take to Alices browser to retrieve all objects and visualize the web page in case it is configured to use a non-persistent connection? Would Alice save time by utilizing a persistent connection with pipelining? How much would save?