[aprssig] FW: Possibility to add-on Server to APRS-IS Tier-2 Pool

[Nagi Punyamurthula]

Thanks for the note; indeed I agree that adding a new server is not necessarily adding more agility to the network.  That begs a side question if a “test bed” environment is available so one could test to see if and how adding a new server to the cluster would impact the overall performance.

From: Jason KG4WSV [mailto:kg4wsv at gmail.com]
Sent: Friday, November 20, 2015 10:40 AM
To: Nagi Punyamurthula <n0agi at n0agi.com>; TAPR APRS Mailing List <aprssig at tapr.org>
Subject: Re: [aprssig] FW: Possibility to add-on Server to APRS-IS Tier-2 Pool


On Fri, Nov 20, 2015 at 9:18 AM, Nagi Punyamurthula via aprssig <aprssig at tapr.org<mailto:aprssig at tapr.org>> wrote:
I wanted to surface this request again and see if anyone had any recommendations or what the request process is to add a server node to the APRS-IS server network

I have no opinion on adding your server to the T2 servers, but I'd like to point out a couple of general things:

proximity in network topology terms does not necessarily relate to geographic proximity.  if link speeds and bandwidths are considered in addition to hop count, latency, etc, in can differ even more.

Depending on the number of servers and number of clients, adding an APRS-IS server can result in a net _increase_ in traffic on the servers you are attempting to "relieve".  For example, if you set up a server you'll probably peer with at least 2 or 3 other servers, and that's a full feed to each.  If the clients that connect to your server only amount to a few filtered feeds, you've created a significant increase in bandwidth on your peer servers in exchange for reducing their load by a few filtered feeds.

"More servers" does not automatically mean "better".

-Jason
kg4wsv

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.tapr.org/pipermail/aprssig_lists.tapr.org/attachments/20151120/d2e19140/attachment.html>