BTCC Controversy: 100 Full Nodes Donated by the Exchange Appear to be Running on Amazon Web Services

07.01.2016

Recently BTCC has launched 100 full nodes across five continents in order to maintain decentralized arrangement of the bitcoin system. Back then, the company explained this move by its wish to keep bitcoin away from any centralization.

Samson Mow, COO at BTCC, stated that the company has launched the nodes in order to help the system operate.

“As a proponent and believer in bitcoin, we see both supporting core development and well as the bitcoin network itself as our duty. The nodes we deployed today will help to preserve the network as the number of full bitcoin nodes continues to decrease,” he said.

Bobby Lee, CEO at BTCC, added:

“We intentionally distributed the nodes we donated in countries that rank low on the full bitcoin node geographic distribution list. We encourage other bitcoin companies to join us in contributing to the bitcoin ecosystem by sponsoring powerful full nodes.”

According to the company’s statements, 21% of the nodes are installed in the USA, 19% both in Germany and Singapore, 13% in China, and 7% each in Australia, Brazil, Ireland, and Japan.

Even though BTCC states its intents are purely altruistic, the cost of running a hundred nodes must be pretty high, so there must be a more commercially-inclined justification for the move. One of possible explanations was provided on Reddit by user jtoomim:

“The problem is that the Chinese government built a system that causes between 1% and 50% packet loss for any packets that cross the country’s border. The exact rate of packet loss varies widely for any given pair of endpoints and from day to day, and it is not dependent on the amount of traffic between the endpoints. This causes TCP to become extremely slow and unreliable. The congestion avoidance algorithms used by TCP assume that packet loss is a sign of congestion and that packet loss goes away if the transmission rate is decreased. Thus, when the Great Firewall drops packets, the TCP algorithms reduce their transmission rate, assuming that that will fix the issue. Since the packet loss rate is not dependent on the amount of traffic, it doesn’t fix the issue, and the packet loss continues. TCP then reduces the traffic flow rate even further. Et cetera. The net result of this is that you often will only get 10 KB/s to 100 KB/s of actual TCP traffic between two peers with 100+ Mbps connections just because they’re on opposite sides of the Great Firewall.”

However, as it turned out, most of those nodes are hosted by Amazon Web Services. This has caused some community members to worry about growing centralization of the system.

Indeed, if a hundred fully validating nodes are in fact operated by the same entity, the system’s decentralization, contrary to the initially stated, may become less unlikely. Some bitcoiners have already said that employment of Amazon Web Services for hosting nodes in order to reinforce the decentralized arrangement does not imply actual deployment of a hundred new nodes, but may consist in purchasing or renting capacities owned by a highly-centralized company.

Reddit user jstolfi commented on possible implications of full non-mining nodes launch:

“Independent “full but non-mining” (FNM) relay nodes have no place in the design of the protocol. Their motivations are dubious, because they have no incentives to be honest — propagate all transactions to the miners, and serve the majority branch of the blockchain. Since they aren’t required to provide proof-of-work, a malicious agent can spawn thousands of them at very little cost. For all that, they make the system less secure, not more.
“What is wrong with the picture is the fact that mining got concentrated into 4-5 large companies. That is definitely not how Satoshi imagined the mining would be. But there seems to be no solution for this problem,”

Found a typo? Highlight text and press CTRL+ENTER

Subscribe to our Newsletter

Tags: Amazon, blockchain, BTCC, nodes