You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Aptos nodes currently use state synchronization (state sync) to stay-up-to-date with the blockchain. When validators execute a new block of transactions, state sync will synchronize the block data to validator fullnodes (VFNs) and public fullnodes (PFNs), ensuring all nodes have the latest blockchain state. However, state sync is not optimal for high-load, low-latency environments. This is because it waits for block data to be processed locally (e.g., executed and committed) before disseminating that data to other nodes in the network. This increases the synchronization time across nodes and negatively impacts the user experience.
To address this, we propose a new synchronization mechanism called Consensus Observer (CO). CO allows Aptos nodes to disseminate block data without having to wait for local processing. For example, when validators first propose and order a block, CO will immediately share that block with the fullnodes (i.e., VFNs and PFNs). This allows the fullnodes to begin processing the block in parallel with the validators, without having to wait for validator execution or commit. This avoids the unnecessary wait times imposed by state sync, and significantly reduces end-to-end transaction latencies at high-load.
AIP Discussion
Aptos nodes currently use state synchronization (state sync) to stay-up-to-date with the blockchain. When validators execute a new block of transactions, state sync will synchronize the block data to validator fullnodes (VFNs) and public fullnodes (PFNs), ensuring all nodes have the latest blockchain state. However, state sync is not optimal for high-load, low-latency environments. This is because it waits for block data to be processed locally (e.g., executed and committed) before disseminating that data to other nodes in the network. This increases the synchronization time across nodes and negatively impacts the user experience.
To address this, we propose a new synchronization mechanism called Consensus Observer (CO). CO allows Aptos nodes to disseminate block data without having to wait for local processing. For example, when validators first propose and order a block, CO will immediately share that block with the fullnodes (i.e., VFNs and PFNs). This allows the fullnodes to begin processing the block in parallel with the validators, without having to wait for validator execution or commit. This avoids the unnecessary wait times imposed by state sync, and significantly reduces end-to-end transaction latencies at high-load.
Read more about it here: https://github.com/aptos-foundation/AIPs/blob/main/aips/aip-93.md
The text was updated successfully, but these errors were encountered: