Author

Topic: Network Size (Read 9097 times)

newbie
Activity: 22
Merit: 0
July 17, 2010, 10:43:20 PM
#9
Based on http://www.alloscomp.com/bitcoin/calculator.php:

1 250 000 khps will have an average generation time of 10 minutes.
  900 000 khps will have a median generation time of 10 minutes.

This thread claims that the average generation speed is about 400 khps, I'm seeing about 1000 (reasonably recent machines).  That would give an estimate between 1000-3000 machines working in the swarm.
newbie
Activity: 22
Merit: 0
July 17, 2010, 08:16:47 PM
#8
To go back to the original posters question: How to estimate the current swarm size?

Seems to me you could get a pretty good estimate of the maximum historical swarm size (in terms of compute power, not nodes) based just by looking at the difficulty number.  As I understand it, the difficulty is scaled up based on how fast the swarm can compute new blocks.

A more accurate and up-to-date estimate could be made based on the time it took for the last few blocks to be solved.

I can't imagine that there is (or could be!) an accurate way of estimating the number of nodes directly.  Too many firewalls, tors, etc.
founder
Activity: 364
Merit: 7423
July 17, 2010, 06:25:16 PM
#7
Version 0.3 was supposed to reduce the number of outgoing connections on non-port forwarded clients from 15 to 8, but I don't think it really happened. I'm not positive if this is the case. Correct me if I'm wrong.
In 0.3.0, the change to 8 only ended up in the Windows version, the other versions still had 15.

Please upgrade to 0.3.2, it's available now.
legendary
Activity: 1246
Merit: 1016
Strength in numbers
July 17, 2010, 05:34:41 PM
#6
I have v3 w/ no forwarding and never get more than 8.
sr. member
Activity: 252
Merit: 268
July 17, 2010, 05:22:09 PM
#5
Version 0.3 was supposed to reduce the number of outgoing connections on non-port forwarded clients from 15 to 8, but I don't think it really happened. I'm not positive if this is the case. Correct me if I'm wrong.
member
Activity: 123
Merit: 15
July 17, 2010, 02:46:48 PM
#4
I am aware that 1 is plenty for proper functioning, but more connections helps prevent the network from becoming segmented.  I was just wondering if there was a limit because I am always at 15 which seemed odd to me.

Anyway in regards to the question I think the protocol itself should support some kind of mechanism like this as it is good to know.  As far as I understand it the IRC server is more of a temporary thing until a more permanent solution is implemented.  This will probably have to happen sooner, rather than later, since the growth of the system will soon overwhelm the IRC server.  It is also somewhat of a single point of failure, even though the clients have other ways of finding eachother losing the IRC server could end up fragmenting the network causing several "islands" of interconnected nodes.  If this islanding persisted for long enough transactions could be lost.

-Buck
legendary
Activity: 860
Merit: 1026
July 17, 2010, 02:37:03 PM
#3
15 connections was the maximum of 0.2 without port forwarding.
but it doesn't matter how much connections you have, 1 is enough, all the others are redundant.
member
Activity: 123
Merit: 15
July 17, 2010, 08:41:52 AM
#2
How is it that you get more than 15, all of my clients exactly level out at 15 and never go higher like there is some hard limit imposed.  Did you have to tweak the source to make that happen?

-Buck
member
Activity: 77
Merit: 10
July 16, 2010, 06:04:44 PM
#1
Is there any way to query the current size of the network, say, in connected clients, actively generating clients, FLOPS, or preferably, all three?  I see that my client is connected, and has been connected, to 15 clients for the past four days....yet somehow I doubt there are only 15 people generating coins.  I'm looking for this in the context of scalability assessments.
Jump to: