2014-05-20
I maintain an OpenPGP key server in a public server pool and regularly receive peering requests from other server operators. To help me better understand the network and make more informed decisions about which requests to respond to, I've compiled a network graph using the monitoring data collected by Kristian Fiskerstrand.
A preliminary look at its degree distribution reveals that most servers are well connected, and several are very much so:
While this is good news regarding the health of the pool, it does pose a challenge to efforts at visualization; the graph is a hairball. In that light, I've taken considerable artistic license to make the network's features more distinguishable:
Hover over a node to clarify its connections.
The size of a node denotes its closeness centrality, a rough predictor of how quickly data might be able to propagate from that server to the rest of the network.
Color is determined by a server's priorities in the three regional pools, Europe (EU), North America (NA), and Oceania (OC):
- Hue indicates the bias between regions, as shown in the key above.
- Saturation corresponds with the strength of that bias. Well balanced servers thus loose any discernible hue.
- Lightness is determined by priority regardless of region. Servers holding a high priority in any region are lighter.
From this graph I can see that my server is relatively well balanced across regions and has a high priority, but has few connections compared to other servers.
I'd hoped that it would be more visually apparent which new connections I should make to strengthen the network, but the graph has turned out to be too complex for me to follow by eye.