A reputation-based dynamic reorganization scheme for blockchain network sharding

Figures & data

Figure 1. System architecture and the process of communicating data.

Figure 2. The expulsion process of nodes.

Figure 3. The working principle of dynamic sharding reconfiguration.

Table

Algorithm 1 Shard Reorganization and Leader Election

Input:

Set of node reputations R,where R=r(v1),r(v2), … ,r(vn)

The k shard C=c1,c2, … ,ck

Output:

Set of shards C

Supervisory Committee, SC

Set of leader nodes L

Initialize an empty set new{}

Extraction process

trusted_shards = shards[:k//2]

regular_shards = shards[k//2:]

for each trusted shard do

Extract α proportion of nodes, including the leader node

Place these nodes within new{}

end for

Add new node allocation process

Add the new nodes to the new{}, set the reputation of nodes to T

Rsort = sort(r(vi))

for each r(vi) in Rsort do

Find a shard Cj that has the minimum reputation

Randomly assign one vi to Cj, Cj = Cj $\cup$ {vi}

end for

Election process

Select the shard with the highest score as the SC

for each shard Ci ∈ C do

Select nodes with scores in the upper 50% as potential leaders.

Randomly select node as l and alter – l

end for

Table 1. PC configurations.

CPU	Intel(R) Core(TM) i7-10700 CPU @ 290GHz
RAM	64GB
OS	Ubuntu server 20.04

Table 2. Performance Capability of experimental node setting.

Performance Capability(C)	Sim-performance
3	100%
2	50%
1	33.3%

Table 3. Solution feature comparison table.

	system setting			Reputation Management
	Allocation	Leader election	Committee reorganisation	behavioural	performance	Node expulsion
Omniledger (Kokoris-Kogias et al., 2018)	Random	Leader	–
Rapidchain (Zamani et al., 2018)	Random	N/A	Cuckoo
TBSD (Yun et al., 2019)	Reputation	Leader	–	√
RDSCM (Shen, 2022)	Reputation	Leader	–	√		√
Proposed	Reputation	Leader+Alter- leader	Reputation	√	√	√

Figure 4. Reputation changes of 6 nodes over 10 epochs.

Figure 5. Comparison of the reputation of different shards.

Figure 6. The standard deviation of the ratio of the number of different malicious nodes.

Figure 7. Failure consensus probability security test.

Figure 8. Collusion attack ratio security test.

Figure 9. Latency testing of the system.

Figure 10. Shard latency of the system.

Kokoris-Kogias, E., Jovanovic, P., Gasser, L., Gailly, N., Syta, E., & Ford, B. (2018). OmniLedger: A secure, scale-out, decentralized ledger via sharding. In 2018 IEEE Symposium on Security and Privacy (SP) (pp. 583–598). https://doi.org/10.1109/SP.2018.000-5.

 Google Scholar

Zamani, M., Movahedi, M., & Raykova, M. (2018). RapidChain: Scaling Blockchain via Full Sharding. In Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security (pp. 931–948). https://doi.org/10.1145/3243734.3243853.

 Google Scholar

Yun, J., Goh, Y., & Chung, J. (2019). Trust-based shard distribution scheme for fault-tolerant shard blockchain networks. IEEE Access, 7, 135164–135175. https://doi.org/10.1109/ACCESS.2019.2942003

 Web of Science ®Google Scholar

Shen, T. (2022). Reputation-Driven dynamic node consensus and reliability sharding model in IoT blockchain. Algorithms, 15. https://doi.org/10.3390/a15020028

 Web of Science ®Google Scholar