Re: More Granularity in the US in the NTP Pool
- From: "Richard B. Gilbert" <rgilbert88@xxxxxxxxxxx>
- Date: Sun, 09 Sep 2007 00:04:13 -0400
Ryan Moore wrote:
The problem with this suggestion is that NTP is mostly effected by the round trip delay on the network. The RTD on the network may have little to do with the physical location of the servers. If you are part of a national ISP, it may be faster to go to a server on the same provider, regardless of the physical location of the server.
Whereas another server in your same town may require going through some large ISP junction point which slows down the RTD.
Even if a server is physically 1 mile from you, the route between you and that server may go through a network junction a long way from you if you use different ISPs.
<snip>
Agreed!! I encountered an extreme case many years ago. I was working at Princeton University at the time. RCA Sarnoff Laboratory was about a mile away. E-mail to Sarnoff Labs, however, traveled via RCA's internet gateway in California! They were 6,000 miles away in network space even though we could almost throw rocks at them!
OTOH, the network delay cannot be LESS than the physical distance even though the converse is not true.
.
- References:
- More Granularity in the US in the NTP Pool
- From: evandro
- Re: More Granularity in the US in the NTP Pool
- From: Ryan Moore
- More Granularity in the US in the NTP Pool
- Prev by Date: Re: More Granularity in the US in the NTP Pool
- Next by Date: Re: More Granularity in the US in the NTP Pool
- Previous by thread: Re: More Granularity in the US in the NTP Pool
- Next by thread: Re: More Granularity in the US in the NTP Pool
- Index(es):
Relevant Pages
|
