Replicated log

As explained in ScyllaDB in Action

A replicated log is an ordered append-only sequence of state-changing entries maintained consistently across multiple nodes.

A replicated log is an ordered, append-only sequence of state-changing entries maintained across multiple nodes. In Raft, a leader receives each request and records it before followers append the same entry, establishing a consistent order for changes such as schema updates. section 8.2.5

A Raft write proceeds in two stages:

• Append the entry to the leader’s local log, reserving its position. section 8.2.5
• Ask other nodes to append the entry to their logs. section 8.2.5
• Commit the entry after a majority of nodes acknowledge the append, making it locked in and persistent. section 8.2.5

Ordering a schema change

If two conflicting schema changes arrive at different nodes, routing requests through one leader places one entry before the other in the replicated log instead of allowing concurrent, conflicting updates. section 8.2.5

Figure 8.9 Raft’s leader appends entries to its log, committing them when a majority of nodes have also acknowledged the append operation.

Trace how a Raft leader appends an entry, replicates it to followers, and commits it after a majority acknowledges the append. section 8.2.5

How ScyllaDB in Action covers it

How deep it goes

Explains leader append, follower replication, and majority commit, but does not detail the complete Raft protocol.

Examples

Uses a leader-and-followers write sequence in which a majority acknowledgment commits a new log entry.

What you can do after

Understand it by tracing how Raft orders changes and commits them after majority replication.

How it connects

Connects Raft with consensus, commit logs, schema changes, topology changes, and cluster-state consistency.

liveBookmanning.com

Read ScyllaDB in Action on liveBook Get the book at manning.com

Related concepts

• Raftpart of · Raft is a distributed consensus algorithm that uses leader election and replicated logs to coordinate nodes and maintain consistent replicated state despite failures.
• logkind of · A log is a record of events, activity, or diagnostic information produced during the operation of a system or process, often ordered by time.
• consensusproduced by · Consensus is sufficient agreement among participants or processes on a shared decision, value, state, assessment, or outcome.
• append-only systempart of · An append-only system preserves existing data and records changes by adding new entries without modifying or removing prior entries.
• transaction logcontrasts with · A transaction log is an ordered record of transactions, transaction-related events, or resulting data changes, maintained as a history and often used for tracking or recovery.
• monitoringcontrasts with · Monitoring is the ongoing observation and measurement of a system’s behavior, state, or operational signals to assess its condition and detect changes or problems.

This explainer was generated by Manning's knowledge graph from the text of ScyllaDB in Action. The section links point to the parts of the book it draws on.