0

I have launched a blockchain testnet in a cluster with different nodes and I have some questions about how it works. If I run more than one geth client(a node in the blockchain) in one node of the cluster:

  • What is the difference between that and run them in different nodes of the cluster?
  • Should all of them mine to obtain the last block added in the blockchain? If one of the geth clients running in a node of the cluster is mining, others geth clients running in the same node also get the last block? (I dont understand very well the interaction between different roles of a node that could be in the blockchain like miner node or full node).
  • Is there any difference sending a transaction between two clients in a node of the cluster and two clients in two different nodes?

E.g enter image description here

1
  • Running nodes in different machines can potentially give you a more realistic simulation of mainnet since you will have some extra latency. But for small cluster it will be almost unnoticeable. Also, running nodes in different machines will make it easier to evaluate the computer power needs / storage needs for a single node. And for security, its easier to hack your network if all nodes are on same computer

  • No, you don't need to have all the nodes mining to relay transactions. Relaying transactions is the most basic role that a node can have and all nodes do this at least.

  • No difference. Eventually the transaction will be relayed to all nodes of the network. If you are concerned with latency, you could relay the transactions at different places of the network. But we are only talking of 100's of milliseconds here..

  • Thanks for the answer. And watching the image, do you know what happens if I am running for example an script in Node 1 which sends a tx from GC1 and another one from GC3 which is in Node2? Is it affecting the latency ? – sapiensapiens Nov 29 '18 at 14:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.