To:
<iljitsch@muada.com>, <dnsop@cafax.se>
From:
<teemu.savolainen@nokia.com>
Date:
Thu, 13 Nov 2003 01:33:09 +0200
content-class:
urn:content-classes:message
Sender:
owner-dnsop@cafax.se
Thread-Index:
AcOpcDEblwOTV04RQ/CeJP01uCAP+QAAuWoQ
Thread-Topic:
DNS discovery
Subject:
RE: DNS discovery
Hi, > adding DHCP for > DNS configuration is exactly that: adding another protocol. > This is bad > because there is now an additional mechanism that can fail > and there is > additional delay in configuring. I agree in this. I'd like to point out that any additional (round trip) delay in configuration phase, especially in wireless (cellular) environments, is _not_ user friendly. Please consider those 200.000.000+ million cellular users opening a network link for browsing etc. purposes who have to wait every time for yet another round trip just because of DNS server address configuration, which could be obtained faster.. The collective amount of consumer time spend for DNS configuration could become large quickly, as wireless (cellular) devices tend to open connections often for short periods of time. Quick example: 100ms extra time spend in every GPRS PDP context opening. 200.000.000 users open 1 PDP context per day for one year: 0.1s*200.000.000*365days = 84490 days = 231 years per every year spent on waiting for server configurations that could be done faster.. In this example. Numbers are estimates, extra time can be bigger, user number more or less.. But this should give some figures. Please correct me if I'm fundamentally mistaken. Additionally if users pay for their connection per every transmitted byte those extra messages cost extra money.. Regards, Teemu #---------------------------------------------------------------------- # To unsubscribe, send a message to <dnsop-request@cafax.se>.