8 Distributed Ledger Technology
411
Delegates are also elected by stakeholders. However, their responsibility is to
maintain the blockchain. For example, delegates can suggest block size changes,
paid incentive, and transaction fee changes. The stakeholders will decide if the
proposed changes should be implemented. Delegates may receive rewards as well.
Energy saving and decentralization promotion are the two main advantages of
DPoS. DPoS needs less energy than PoW because witnesses generate blocks based
on specific time schedules, rather than competing with each other to add blocks.
The computing hardware requirement is no longer as demanding as PoW as well.
In addition, greater decentralization is achieved in DPoS because its consensus
mechanism allows stakeholders to choose suitable witnesses to validate transactions.
The main disadvantage of the DPoS consensus mechanism is that it can never
achieve full decentralization, although decentralization can be increased by having
more witnesses validate blocks, due to scalability constraints.
8.3.4.4 Practical Byzantine Fault Tolerance (PBFT)
The practical Byzantine Fault Tolerance (PBFT) consensus algorithm [13] is another
popular consensus algorithm used in blockchains. PBFT enables a blockchain to
tolerate Byzantine faults, i.e., defend against attacks from malicious nodes. The
algorithm is designed to work in asynchronous systems. PBFT has low overhead
time and low latency.
In PBFT, all nodes of a blockchain are organized into a sequence. A specific node
is designated as the leader node. Other nodes are designated as backup nodes. When
a node sends out a message, the rest of the nodes will exchange information with
each other to validate the message in case it is tampered during transmission. It is
expected that the good nodes will reach an agreement on the state of the blockchain
through majority.
Each round (called view) of the PBFT works as follows:
1. A client sends a request to the leader node.
2. The leader node broadcasts the request to backup nodes.
3. The backup nodes execute the request and send a response to the client.
4. The client waits to receive f + 1 node responses with the same result which will
be used as the result of the operation, where f represents the maximum number
of potentially faulty nodes.
To secure its role, the leader node may be changed in a round-robin fashion
during every view. The leader node can even be replaced if it does not broadcast a
request after a given time interval. The majority of good nodes also have the power
to identify a faulty leader node and replace it with the next leader.
411
Delegates are also elected by stakeholders. However, their responsibility is to
maintain the blockchain. For example, delegates can suggest block size changes,
paid incentive, and transaction fee changes. The stakeholders will decide if the
proposed changes should be implemented. Delegates may receive rewards as well.
Energy saving and decentralization promotion are the two main advantages of
DPoS. DPoS needs less energy than PoW because witnesses generate blocks based
on specific time schedules, rather than competing with each other to add blocks.
The computing hardware requirement is no longer as demanding as PoW as well.
In addition, greater decentralization is achieved in DPoS because its consensus
mechanism allows stakeholders to choose suitable witnesses to validate transactions.
The main disadvantage of the DPoS consensus mechanism is that it can never
achieve full decentralization, although decentralization can be increased by having
more witnesses validate blocks, due to scalability constraints.
8.3.4.4 Practical Byzantine Fault Tolerance (PBFT)
The practical Byzantine Fault Tolerance (PBFT) consensus algorithm [13] is another
popular consensus algorithm used in blockchains. PBFT enables a blockchain to
tolerate Byzantine faults, i.e., defend against attacks from malicious nodes. The
algorithm is designed to work in asynchronous systems. PBFT has low overhead
time and low latency.
In PBFT, all nodes of a blockchain are organized into a sequence. A specific node
is designated as the leader node. Other nodes are designated as backup nodes. When
a node sends out a message, the rest of the nodes will exchange information with
each other to validate the message in case it is tampered during transmission. It is
expected that the good nodes will reach an agreement on the state of the blockchain
through majority.
Each round (called view) of the PBFT works as follows:
1. A client sends a request to the leader node.
2. The leader node broadcasts the request to backup nodes.
3. The backup nodes execute the request and send a response to the client.
4. The client waits to receive f + 1 node responses with the same result which will
be used as the result of the operation, where f represents the maximum number
of potentially faulty nodes.
To secure its role, the leader node may be changed in a round-robin fashion
during every view. The leader node can even be replaced if it does not broadcast a
request after a given time interval. The majority of good nodes also have the power
to identify a faulty leader node and replace it with the next leader.