Recently there have been some discussion around problem of nodes having to download the entire operation tree (which could be slow) in order to catch up.
If you're interested in this problem you can take a look and chime in here:
https://github.com/ipfs/dynamic-data-and-capabilities/issues/14
State-based CRDTs ensure convergence through disseminating the entire state, that may be large, and merging it to other replicas; whereas operation-based CRDTs disseminate operations (i.e., small states) assuming an exactly-once reliable dissemination layer.
Operation-based CRDTs disseminate operations (i.e., small states) assuming an exactly-once reliable dissemination layer.
This paper introduces Delta State Conflict-Free Replicated Data Types (未-CRDT) that can achieve the best of both worlds: small messages with an incremental nature, as in operation-based CRDTs, disseminated over unreliable communi- cation channels, as in traditional state-based CRDTs.
Benefits of delta CRDTs over operation-based CRDTs:
Link to the paper: https://www.sciencedirect.com/science/article/pii/S0743731517302332
Another paper brought to my attention by @diasdavid:
"Efficient Synchronization of State-based CRDTs": https://arxiv.org/abs/1803.02750
Missed that, thank you @sunny-g !
Most helpful comment
未-CRDT
State-based CRDTs ensure convergence through disseminating the entire state, that may be large, and merging it to other replicas; whereas operation-based CRDTs disseminate operations (i.e., small states) assuming an exactly-once reliable dissemination layer.
Operation-based CRDTs disseminate operations (i.e., small states) assuming an exactly-once reliable dissemination layer.
This paper introduces Delta State Conflict-Free Replicated Data Types (未-CRDT) that can achieve the best of both worlds: small messages with an incremental nature, as in operation-based CRDTs, disseminated over unreliable communi- cation channels, as in traditional state-based CRDTs.
Benefits of delta CRDTs over operation-based CRDTs: