8

IOTA is scalable in the sense that the more transactions are being made, the faster transactions will be confirmed.

But what I fail to understand (for any crypto-currency really) is that if every node needs to receive all transactions how will IOTA be able to scale beyond the number of transactions that can be handled by a single node?

Secondly if IOTA really becomes a backbone for the Internet of Things, won't the tangle grow so fast that you need to snapshot almost continuously?

|improve this question|||||
  • Valid question. I asked a similar one here. My understanding that in the future some kind of swarm nodes should be implemented, so every node will validate just part of the transactions in the system (small part), not all of them. – alexpods Feb 4 '18 at 11:45
2

Yes, Fullnodes have to sync and store every single transactions and with Fullnodes it is not possible to scale unlimited. The current Network is build by Fullnodes and is not able to scale.

The solution of this bottleneck is the future implementation of swarmnodes as mentioned in the Roadmap. Swarmnodes are Nodes which don't sync and store every single transaction. Actually there are/were some Swarmnodes already, but they are still under R&D by Come-from-Beyond.

Some Speculation:

One of the mayor topics in Swarmnode R&D might be the optimal size of the synced neighborhood.

Problem, which might occure:

You can't spend funds in areas that don't know your balance.

|improve this answer|||||
  • This is the answer I was looking for. Sounds very tricky though and probably will probably impose several limits. – Stefan Feb 18 '18 at 12:09
  • Following limits can be overcome by sharing snapshots over the entire network or similar solutions. So there is a lot of research been done at the moment and we have to be patient. – Tobi MZ Feb 18 '18 at 12:19
2

Right now the network is being limited by the coordinator. My following answer is made with the assumption that the coordinator has been removed or at least decentralized into the network.

We need enough full nodes and suitable peer discovery algorithms to make sure no single full node can get congested. If this can be satisfied then theoretically the Tangle is infinitely scalable. The more transactions the merrier.

Automated snapshotting is in the roadmap. In the future, there will not be a need to perform snapshots manually. The process will be automated.

|improve this answer|||||
  • 1
    The question was that even with enough full nodes and suitable peer discovery algorithms still EVERY SINGLE FULL NODE in the system has to validate EVERY SINGLE TRANSACTION in the system, which means, despite how many nodes are deployed, still the scalability of the whole network is limited by the capacity of just ONE node in that network. – alexpods Feb 4 '18 at 11:39
  • 1
    To my understanding, full nodes do not have to validate every single transaction. They only validate the unconfirmed transactions that they receives. But it is true that every full node will have to receive information of all confirmed transactions in the Tangle, as of the current state of implementation. This can be a bottleneck of the network at the moment. However, there is active R&D going on in the area of swarm clients and sharding, which will tackle this issue. – Bayequentist Feb 4 '18 at 20:15

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.