Qtum has proposed a new consensus algorithm, called SCAR, that could massively reduce transaction times. It would additionally lighten demands on the blockchain’s underlying network.
“SCAR can find a balance between user experience and resource saving, which is more efficient and flexible than other consensus algorithms,” reads the proposal.
SCAR would essentially allow for block intervals to be adjusted based on the transaction load that a network is experiencing. How will SCAR work, and how is it different from other mechanisms?
On the blockchain, each block carries several transactions, and a block must be confirmed in order for its transactions to be completed. Block intervals are the spans of time between those confirmations.
All blockchains rely on an underlying network to perform confirmations, but each can rely on a different consensus mechanism. Some blockchains rely on a network made up of many miners, while others rely on a consortium of supernodes or elected block producers. Different consensus mechanisms generally produce different block intervals.
Proof-of-work (PoW) systems like Bitcoin rely on a large number of miners. Although these systems distribute resource demands widely across a large network without burdening any miner too much, they often have lengthy block intervals–Bitcoin’s block interval, for example, is 10 minutes long. This slows down payments and is not optimal for a widely-used payment network.
Meanwhile, more recent “consortium” consensus mechanisms, such as Qtum’s Aura and EOS’s DPOS, rely on a smaller number of nodes that are responsible for confirming blocks. These systems demand heavy resources from a few powerful nodes, who must dedicate plenty of bandwidth and disk space. This approach does, however, reduce block intervals to split-second durations.
Qtum’s goal is to find a middle ground: to improve transaction speed while remaining sensitive to resource demands. With this in mind, Qtum is proposing a system that would constantly change block intervals on its blockchain.
Right now, many blockchains adjust their difficulty in order to ensure that their block intervals remain constant. This ensures that the network performs consistently, even if the number of miners or staked tokens drops.
This keeps transaction times stable in the face of network fluctuations, but not in the face of high transaction volumes. SCAR would solve this problem by calculating a block interval based on the network’s transaction load.
After that, each supernode would attempt to produce the highest-priority block based on time-related parameters. Once a block is produced, new supernodes would be voted in and the cycle would repeat.
SCAR would provide various benefits. In addition to quick transaction times for users, it would put less demand on supernodes, who would otherwise provide bandwidth and disk space needlessly:
“[SCAR] can effectively avoid wasting precious storage resources when the transaction volume is small; also it can increase the block generation rate when the transaction volume is large.”
Qtum’s concept is, furthermore, generally applicable and “not limited to the one proposed in [the] paper,” which means that it could potentially be used in any blockchain and innovated upon.