I will abstain on this wrangle of "when", 

Instead I'd like to address some of the network topology health issues that's been brought up in this debate.

Due to how blocks are being broadcast by miners at the moment, it is not difficult to find the origin node of these blocks. These more influential origin nodes are a minority, about <100 out of ~6000, <2%. These data points are important to certain attack vectors. It is highly recommended that pools adopt broadcast logic that rotates broadcasting nodes and increase their node count.. Eloipool has this implanted for those seeking to adopt/see it in action in the wild.

China is a particular worse-case due to the sporadic nature of their internet infrastructure, especially connecting from/to outside of gfw, on a average node-walk I can get up to a 10% difference while I know for a fact some of the nodes shown to be down are up.

In F2Pool's case, I see 6 replay nodes, I don't know if that's enough or that's all the nodes F2Pool runs, but it may be beneficial to set up multi-homing with shadowsocks over mptcp to increase the stability. also see if you can get a CERNET connection to be part of your rotations since their backbone is quite good.

comments, question and grievances welcome.

On Sat, May 30, 2015 at 9:31 PM, Chun Wang <1240902@gmail.com> wrote:
On Sat, May 30, 2015 at 9:57 PM, Gavin Andresen <gavinandresen@gmail.com> wrote:
>> Bad miners could attack us and the network with artificial
>> big blocks.
>
>
> How?
>
> I ran some simulations, and I could not find a network topology where a big
> miner producing big blocks could cause a loss of profit to another miner
> (big or small) producing smaller blocks:
>
> http://gavinandresen.ninja/are-bigger-blocks-better-for-bigger-miners
>
> (the 0.3% advantage I DID find was for the situation where EVERYBODY was
> producing big blocks).

If someone propagate a 20MB block, it will take at best 6 seconds for
us to receive to verify it at current configuration, result of one
percent orphan rate increase. Or, we can mine the next block only on
the previous block's header, in this case, the network would see many
more transaction-less blocks.

Our orphan rate is about 0.5% over the past few months. If the network
floods 20MB blocks, it can be well above 2%. Besides bandwidth, A 20MB
block could contain an average of 50000 transactions, hundred of
thousands of sigops, Do you have an estimate how long it takes on the
submitblock rpccall?

For references, our 30Mbps bandwidth in Beijing costs us 1350 dollars
per month. We also use Aliyun and Linode cloud services for block
propagation. As of May 2015, the price is 0.13 U.S. dollars per GB for
100Mbps connectivity at Aliyun. For a single cross-border TCP
connection, it would be certainly far slower than 12.5 MB/s.

I think we can accept 5MB block at most.

(sorry forgot to cc to the mailing list)

------------------------------------------------------------------------------
_______________________________________________
Bitcoin-development mailing list
Bitcoin-development@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bitcoin-development



--
Yifu Guo
"Life is an everlasting self-improvement."