Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo merged PR #1338: URL: https://github.com/apache/ratis/pull/1338 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on PR #1338: URL: https://github.com/apache/ratis/pull/1338#issuecomment-3837096197 "Split-brain" behavior revised as suggested, and squashed all commits to this point. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2755815651 ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to Apache Ratis + +Previous: [Integration](integration.md) | Top:[Overview of Raft and Ratis](index-v2.md) + +## Section 4: Operations and Management + +* [Snapshots](#snapshots---managing-growth-and-recovery) +* [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) + +### Snapshots - Managing Growth and Recovery + +Snapshots are a point-in-time representation of your state machine's complete state, along with +metadata about which log entries are included in that state. They prevent the log from growing +without bound and enable efficient recovery and catch-up for peers that have fallen behind. + +The snapshot includes the actual application state, the term-index of the last log entry that +contributed to this state, and the Raft group configuration at the time the snapshot was taken. + +Without snapshots, the Raft log would grow indefinitely, eventually consuming all available +storage. Crashed peers would need to replay potentially millions of log entries to catch up, +dramatically slowing recovery. New peers joining an established group would need to process the +entire history of the group, which could take hours or days for active systems. + + Creating Snapshots + +Snapshots can be created automatically when the log grows beyond a certain size, manually +triggered through the admin API, or sent by the leader to peers that are far behind instead of +replaying thousands of log entries. + +When your state machine's `takeSnapshot` method is called, it needs to create a consistent view +of your application state. This might involve pausing writes, creating a database transaction, +or using copy-on-write data structures. The method must serialize state by writing it to durable +storage in a format that can be read back later, record which term-index the snapshot represents, +and return the log index so Ratis can safely discard older log entries. + +Different applications will have different strategies for snapshot creation. A stop-the-world +approach pauses all operations while creating the snapshot: simple but impacts availability. +Copy-on-write uses data structures that support efficient point-in-time copies. Database +transactions can create consistent snapshots if your state is in a database. Some storage +engines support checkpointing to leverage native snapshot capabilities. + + Snapshot Installation and Recovery + +When a peer needs to catch up using a snapshot, it receives the snapshot data from the leader or +loads it from local storage. The state machine is paused to prevent conflicts during restoration, +the snapshot data is loaded replacing any existing state, and the state machine resumes normal +operation by replaying any log entries that occurred after the snapshot. + +Your state machine's `reinitialize` method is responsible for loading snapshots during startup by +loading the latest snapshot if available, with the Raft layer replaying any log entries after +the snapshot. + + Designing Snapshot-Friendly State Machines + +When designing your state machine, ensure your state can be efficiently serialized and +deserialized, avoiding complex object graphs that are difficult to serialize. For very large +state machines, consider whether you can implement incremental snapshots that only capture +changes since the last snapshot. + +If your state machine maintains state in external systems, ensure your snapshot process captures +this external state consistently. Regularly test your snapshot and recovery process to ensure it +works correctly under various failure scenarios. + +### Leadership and Fault Tolerance + +Leadership in Ratis is both simpler and more complex than it might initially appear. Ratis +handles all the mechanics of leader election and failover automatically, but your application +needs to handle leadership changes robustly. + + Leadership and Automatic Election + +In Raft, the leader is the only server that can accept write requests and decide the order of +operations in the log. This centralized decision-making enables Raft to provide strong +consistency guarantees. Leadership is temporary and can change at any time due to failures, +network partitions, or normal operational events. + +When a Raft group starts up, or when the current leader fails, the remaining servers +automatically elect a new leader through a voting process. This process uses randomized timeouts +to prevent split votes and ensures that only servers with up-to-date logs can become leaders. +This happens entirely within Ratis without any intervention from your application code. + + Leadership and Client Behavior + +From a client perspective, leadership changes are largely transparent. Clients can send requests +to any se
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2755794405 ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to Apache Ratis + +Previous: [Integration](integration.md) | Top:[Overview of Raft and Ratis](index-v2.md) + +## Section 4: Operations and Management + +* [Snapshots](#snapshots---managing-growth-and-recovery) +* [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) + +### Snapshots - Managing Growth and Recovery + +Snapshots are a point-in-time representation of your state machine's complete state, along with +metadata about which log entries are included in that state. They prevent the log from growing +without bound and enable efficient recovery and catch-up for peers that have fallen behind. + +The snapshot includes the actual application state, the term-index of the last log entry that +contributed to this state, and the Raft group configuration at the time the snapshot was taken. + +Without snapshots, the Raft log would grow indefinitely, eventually consuming all available +storage. Crashed peers would need to replay potentially millions of log entries to catch up, +dramatically slowing recovery. New peers joining an established group would need to process the +entire history of the group, which could take hours or days for active systems. + + Creating Snapshots + +Snapshots can be created automatically when the log grows beyond a certain size, manually +triggered through the admin API, or sent by the leader to peers that are far behind instead of +replaying thousands of log entries. + +When your state machine's `takeSnapshot` method is called, it needs to create a consistent view +of your application state. This might involve pausing writes, creating a database transaction, +or using copy-on-write data structures. The method must serialize state by writing it to durable +storage in a format that can be read back later, record which term-index the snapshot represents, +and return the log index so Ratis can safely discard older log entries. + +Different applications will have different strategies for snapshot creation. A stop-the-world +approach pauses all operations while creating the snapshot: simple but impacts availability. +Copy-on-write uses data structures that support efficient point-in-time copies. Database +transactions can create consistent snapshots if your state is in a database. Some storage +engines support checkpointing to leverage native snapshot capabilities. + + Snapshot Installation and Recovery + +When a peer needs to catch up using a snapshot, it receives the snapshot data from the leader or +loads it from local storage. The state machine is paused to prevent conflicts during restoration, +the snapshot data is loaded replacing any existing state, and the state machine resumes normal +operation by replaying any log entries that occurred after the snapshot. + +Your state machine's `reinitialize` method is responsible for loading snapshots during startup by +loading the latest snapshot if available, with the Raft layer replaying any log entries after +the snapshot. + + Designing Snapshot-Friendly State Machines + +When designing your state machine, ensure your state can be efficiently serialized and +deserialized, avoiding complex object graphs that are difficult to serialize. For very large +state machines, consider whether you can implement incremental snapshots that only capture +changes since the last snapshot. + +If your state machine maintains state in external systems, ensure your snapshot process captures +this external state consistently. Regularly test your snapshot and recovery process to ensure it +works correctly under various failure scenarios. + +### Leadership and Fault Tolerance + +Leadership in Ratis is both simpler and more complex than it might initially appear. Ratis +handles all the mechanics of leader election and failover automatically, but your application +needs to handle leadership changes robustly. + + Leadership and Automatic Election + +In Raft, the leader is the only server that can accept write requests and decide the order of +operations in the log. This centralized decision-making enables Raft to provide strong +consistency guarantees. Leadership is temporary and can change at any time due to failures, +network partitions, or normal operational events. + +When a Raft group starts up, or when the current leader fails, the remaining servers +automatically elect a new leader through a voting process. This process uses randomized timeouts +to prevent split votes and ensures that only servers with up-to-date logs can become leaders. +This happens entirely within Ratis without any intervention from your application code. + + Leadership and Client Behavior + +From a client perspective, leadership changes are largely transparent. Clients can send requests +to any serv
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2755794405 ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to Apache Ratis + +Previous: [Integration](integration.md) | Top:[Overview of Raft and Ratis](index-v2.md) + +## Section 4: Operations and Management + +* [Snapshots](#snapshots---managing-growth-and-recovery) +* [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) + +### Snapshots - Managing Growth and Recovery + +Snapshots are a point-in-time representation of your state machine's complete state, along with +metadata about which log entries are included in that state. They prevent the log from growing +without bound and enable efficient recovery and catch-up for peers that have fallen behind. + +The snapshot includes the actual application state, the term-index of the last log entry that +contributed to this state, and the Raft group configuration at the time the snapshot was taken. + +Without snapshots, the Raft log would grow indefinitely, eventually consuming all available +storage. Crashed peers would need to replay potentially millions of log entries to catch up, +dramatically slowing recovery. New peers joining an established group would need to process the +entire history of the group, which could take hours or days for active systems. + + Creating Snapshots + +Snapshots can be created automatically when the log grows beyond a certain size, manually +triggered through the admin API, or sent by the leader to peers that are far behind instead of +replaying thousands of log entries. + +When your state machine's `takeSnapshot` method is called, it needs to create a consistent view +of your application state. This might involve pausing writes, creating a database transaction, +or using copy-on-write data structures. The method must serialize state by writing it to durable +storage in a format that can be read back later, record which term-index the snapshot represents, +and return the log index so Ratis can safely discard older log entries. + +Different applications will have different strategies for snapshot creation. A stop-the-world +approach pauses all operations while creating the snapshot: simple but impacts availability. +Copy-on-write uses data structures that support efficient point-in-time copies. Database +transactions can create consistent snapshots if your state is in a database. Some storage +engines support checkpointing to leverage native snapshot capabilities. + + Snapshot Installation and Recovery + +When a peer needs to catch up using a snapshot, it receives the snapshot data from the leader or +loads it from local storage. The state machine is paused to prevent conflicts during restoration, +the snapshot data is loaded replacing any existing state, and the state machine resumes normal +operation by replaying any log entries that occurred after the snapshot. + +Your state machine's `reinitialize` method is responsible for loading snapshots during startup by +loading the latest snapshot if available, with the Raft layer replaying any log entries after +the snapshot. + + Designing Snapshot-Friendly State Machines + +When designing your state machine, ensure your state can be efficiently serialized and +deserialized, avoiding complex object graphs that are difficult to serialize. For very large +state machines, consider whether you can implement incremental snapshots that only capture +changes since the last snapshot. + +If your state machine maintains state in external systems, ensure your snapshot process captures +this external state consistently. Regularly test your snapshot and recovery process to ensure it +works correctly under various failure scenarios. + +### Leadership and Fault Tolerance + +Leadership in Ratis is both simpler and more complex than it might initially appear. Ratis +handles all the mechanics of leader election and failover automatically, but your application +needs to handle leadership changes robustly. + + Leadership and Automatic Election + +In Raft, the leader is the only server that can accept write requests and decide the order of +operations in the log. This centralized decision-making enables Raft to provide strong +consistency guarantees. Leadership is temporary and can change at any time due to failures, +network partitions, or normal operational events. + +When a Raft group starts up, or when the current leader fails, the remaining servers +automatically elect a new leader through a voting process. This process uses randomized timeouts +to prevent split votes and ensures that only servers with up-to-date logs can become leaders. +This happens entirely within Ratis without any intervention from your application code. + + Leadership and Client Behavior + +From a client perspective, leadership changes are largely transparent. Clients can send requests +to any serv
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on PR #1338: URL: https://github.com/apache/ratis/pull/1338#issuecomment-3837017855 Thanks. With the exception of my question above (https://github.com/apache/ratis/pull/1338#discussion_r2749937107), I've incorporated your feedback. I also moved my working copy of index.md to index.md and squashed the commits thus far. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2755638516 ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to Apache Ratis + +Previous: [Integration](integration.md) | Top:[Overview of Raft and Ratis](index-v2.md) + +## Section 4: Operations and Management + +* [Snapshots](#snapshots---managing-growth-and-recovery) +* [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) + +### Snapshots - Managing Growth and Recovery + +Snapshots are a point-in-time representation of your state machine's complete state, along with +metadata about which log entries are included in that state. They prevent the log from growing +without bound and enable efficient recovery and catch-up for peers that have fallen behind. + +The snapshot includes the actual application state, the term-index of the last log entry that +contributed to this state, and the Raft group configuration at the time the snapshot was taken. + +Without snapshots, the Raft log would grow indefinitely, eventually consuming all available +storage. Crashed peers would need to replay potentially millions of log entries to catch up, +dramatically slowing recovery. New peers joining an established group would need to process the +entire history of the group, which could take hours or days for active systems. + + Creating Snapshots + +Snapshots can be created automatically when the log grows beyond a certain size, manually +triggered through the admin API, or sent by the leader to peers that are far behind instead of +replaying thousands of log entries. + +When your state machine's `takeSnapshot` method is called, it needs to create a consistent view +of your application state. This might involve pausing writes, creating a database transaction, +or using copy-on-write data structures. The method must serialize state by writing it to durable +storage in a format that can be read back later, record which term-index the snapshot represents, +and return the log index so Ratis can safely discard older log entries. + +Different applications will have different strategies for snapshot creation. A stop-the-world +approach pauses all operations while creating the snapshot: simple but impacts availability. +Copy-on-write uses data structures that support efficient point-in-time copies. Database +transactions can create consistent snapshots if your state is in a database. Some storage +engines support checkpointing to leverage native snapshot capabilities. + + Snapshot Installation and Recovery + +When a peer needs to catch up using a snapshot, it receives the snapshot data from the leader or +loads it from local storage. The state machine is paused to prevent conflicts during restoration, +the snapshot data is loaded replacing any existing state, and the state machine resumes normal +operation by replaying any log entries that occurred after the snapshot. + +Your state machine's `reinitialize` method is responsible for loading snapshots during startup by +loading the latest snapshot if available, with the Raft layer replaying any log entries after +the snapshot. + + Designing Snapshot-Friendly State Machines + +When designing your state machine, ensure your state can be efficiently serialized and +deserialized, avoiding complex object graphs that are difficult to serialize. For very large +state machines, consider whether you can implement incremental snapshots that only capture +changes since the last snapshot. + +If your state machine maintains state in external systems, ensure your snapshot process captures +this external state consistently. Regularly test your snapshot and recovery process to ensure it +works correctly under various failure scenarios. + +### Leadership and Fault Tolerance + +Leadership in Ratis is both simpler and more complex than it might initially appear. Ratis +handles all the mechanics of leader election and failover automatically, but your application +needs to handle leadership changes robustly. + + Leadership and Automatic Election + +In Raft, the leader is the only server that can accept write requests and decide the order of +operations in the log. This centralized decision-making enables Raft to provide strong +consistency guarantees. Leadership is temporary and can change at any time due to failures, +network partitions, or normal operational events. + +When a Raft group starts up, or when the current leader fails, the remaining servers +automatically elect a new leader through a voting process. This process uses randomized timeouts +to prevent split votes and ensures that only servers with up-to-date logs can become leaders. +This happens entirely within Ratis without any intervention from your application code. + + Leadership and Client Behavior + +From a client perspective, leadership changes are largely transparent. Clients can send requests +to any se
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2755586054 ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to Apache Ratis + +Previous: [Integration](integration.md) | Top:[Overview of Raft and Ratis](index-v2.md) + +## Section 4: Operations and Management + +* [Snapshots](#snapshots---managing-growth-and-recovery) +* [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) + +### Snapshots - Managing Growth and Recovery + +Snapshots are a point-in-time representation of your state machine's complete state, along with +metadata about which log entries are included in that state. They prevent the log from growing +without bound and enable efficient recovery and catch-up for peers that have fallen behind. + +The snapshot includes the actual application state, the term-index of the last log entry that +contributed to this state, and the Raft group configuration at the time the snapshot was taken. + +Without snapshots, the Raft log would grow indefinitely, eventually consuming all available +storage. Crashed peers would need to replay potentially millions of log entries to catch up, +dramatically slowing recovery. New peers joining an established group would need to process the +entire history of the group, which could take hours or days for active systems. + + Creating Snapshots + +Snapshots can be created automatically when the log grows beyond a certain size, manually +triggered through the admin API, or sent by the leader to peers that are far behind instead of +replaying thousands of log entries. + +When your state machine's `takeSnapshot` method is called, it needs to create a consistent view +of your application state. This might involve pausing writes, creating a database transaction, +or using copy-on-write data structures. The method must serialize state by writing it to durable +storage in a format that can be read back later, record which term-index the snapshot represents, +and return the log index so Ratis can safely discard older log entries. + +Different applications will have different strategies for snapshot creation. A stop-the-world +approach pauses all operations while creating the snapshot: simple but impacts availability. +Copy-on-write uses data structures that support efficient point-in-time copies. Database +transactions can create consistent snapshots if your state is in a database. Some storage +engines support checkpointing to leverage native snapshot capabilities. + + Snapshot Installation and Recovery + +When a peer needs to catch up using a snapshot, it receives the snapshot data from the leader or +loads it from local storage. The state machine is paused to prevent conflicts during restoration, +the snapshot data is loaded replacing any existing state, and the state machine resumes normal +operation by replaying any log entries that occurred after the snapshot. + +Your state machine's `reinitialize` method is responsible for loading snapshots during startup by +loading the latest snapshot if available, with the Raft layer replaying any log entries after +the snapshot. + + Designing Snapshot-Friendly State Machines + +When designing your state machine, ensure your state can be efficiently serialized and +deserialized, avoiding complex object graphs that are difficult to serialize. For very large +state machines, consider whether you can implement incremental snapshots that only capture +changes since the last snapshot. + +If your state machine maintains state in external systems, ensure your snapshot process captures +this external state consistently. Regularly test your snapshot and recovery process to ensure it +works correctly under various failure scenarios. + +### Leadership and Fault Tolerance + +Leadership in Ratis is both simpler and more complex than it might initially appear. Ratis +handles all the mechanics of leader election and failover automatically, but your application +needs to handle leadership changes robustly. Review Comment: Thanks: I think that's a better phrasing. Before I read your full comment just now I was going to raise the scenario of client that strongly prefers follower reads needed some kind of application logic to decide what to do if a follower is elected leader, but agree the phrasing as written over-states the need. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on PR #1338: URL: https://github.com/apache/ratis/pull/1338#issuecomment-3836779070 > BTW, have you used any AI generative tool to create the doc? Yes, definitely: I've been using Amazon Q CLI to analyze the Ratis and Ozone codebases, as well as public documentation, and generate the initial document drafts, as well as helping with organization. I've personally reviewed every line of documentation and made corrections/adjustments where I felt they were needed, so I own any errors. Reviewing the rest of your feedback now: thanks. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2749933840 ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to Apache Ratis + +Previous: [Integration](integration.md) | Top:[Overview of Raft and Ratis](index-v2.md) + +## Section 4: Operations and Management + +* [Snapshots](#snapshots---managing-growth-and-recovery) +* [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) + +### Snapshots - Managing Growth and Recovery + +Snapshots are a point-in-time representation of your state machine's complete state, along with +metadata about which log entries are included in that state. They prevent the log from growing +without bound and enable efficient recovery and catch-up for peers that have fallen behind. + +The snapshot includes the actual application state, the term-index of the last log entry that +contributed to this state, and the Raft group configuration at the time the snapshot was taken. + +Without snapshots, the Raft log would grow indefinitely, eventually consuming all available +storage. Crashed peers would need to replay potentially millions of log entries to catch up, +dramatically slowing recovery. New peers joining an established group would need to process the +entire history of the group, which could take hours or days for active systems. + + Creating Snapshots + +Snapshots can be created automatically when the log grows beyond a certain size, manually +triggered through the admin API, or sent by the leader to peers that are far behind instead of +replaying thousands of log entries. + +When your state machine's `takeSnapshot` method is called, it needs to create a consistent view +of your application state. This might involve pausing writes, creating a database transaction, +or using copy-on-write data structures. The method must serialize state by writing it to durable +storage in a format that can be read back later, record which term-index the snapshot represents, +and return the log index so Ratis can safely discard older log entries. + +Different applications will have different strategies for snapshot creation. A stop-the-world +approach pauses all operations while creating the snapshot: simple but impacts availability. +Copy-on-write uses data structures that support efficient point-in-time copies. Database +transactions can create consistent snapshots if your state is in a database. Some storage +engines support checkpointing to leverage native snapshot capabilities. + + Snapshot Installation and Recovery + +When a peer needs to catch up using a snapshot, it receives the snapshot data from the leader or +loads it from local storage. The state machine is paused to prevent conflicts during restoration, +the snapshot data is loaded replacing any existing state, and the state machine resumes normal +operation by replaying any log entries that occurred after the snapshot. + +Your state machine's `reinitialize` method is responsible for loading snapshots during startup by +loading the latest snapshot if available, with the Raft layer replaying any log entries after +the snapshot. + + Designing Snapshot-Friendly State Machines + +When designing your state machine, ensure your state can be efficiently serialized and +deserialized, avoiding complex object graphs that are difficult to serialize. For very large +state machines, consider whether you can implement incremental snapshots that only capture +changes since the last snapshot. + +If your state machine maintains state in external systems, ensure your snapshot process captures +this external state consistently. Regularly test your snapshot and recovery process to ensure it +works correctly under various failure scenarios. + +### Leadership and Fault Tolerance + +Leadership in Ratis is both simpler and more complex than it might initially appear. Ratis +handles all the mechanics of leader election and failover automatically, but your application +needs to handle leadership changes robustly. Review Comment: Why "your application needs to handle leadership changes robustly" ? Only the applications want to know who is the leader need to handle leadership changes. If an application does not care about who is the leader, then it does not need to do anything. The first sentence is a kind of a personal comment. Let's remove it. I suggest to rewrite this paragraph as below: ```md Ratis handles all the mechanics of leader election and failover automatically. If your application does not care about who is the leader, then it does not need to do anything. Otherwise, your application can optionally observe leadership change and react accordingly; see [State Machine Leadership Events](#State-Machine-Leadership-Events). ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2749933840 ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to Apache Ratis + +Previous: [Integration](integration.md) | Top:[Overview of Raft and Ratis](index-v2.md) + +## Section 4: Operations and Management + +* [Snapshots](#snapshots---managing-growth-and-recovery) +* [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) + +### Snapshots - Managing Growth and Recovery + +Snapshots are a point-in-time representation of your state machine's complete state, along with +metadata about which log entries are included in that state. They prevent the log from growing +without bound and enable efficient recovery and catch-up for peers that have fallen behind. + +The snapshot includes the actual application state, the term-index of the last log entry that +contributed to this state, and the Raft group configuration at the time the snapshot was taken. + +Without snapshots, the Raft log would grow indefinitely, eventually consuming all available +storage. Crashed peers would need to replay potentially millions of log entries to catch up, +dramatically slowing recovery. New peers joining an established group would need to process the +entire history of the group, which could take hours or days for active systems. + + Creating Snapshots + +Snapshots can be created automatically when the log grows beyond a certain size, manually +triggered through the admin API, or sent by the leader to peers that are far behind instead of +replaying thousands of log entries. + +When your state machine's `takeSnapshot` method is called, it needs to create a consistent view +of your application state. This might involve pausing writes, creating a database transaction, +or using copy-on-write data structures. The method must serialize state by writing it to durable +storage in a format that can be read back later, record which term-index the snapshot represents, +and return the log index so Ratis can safely discard older log entries. + +Different applications will have different strategies for snapshot creation. A stop-the-world +approach pauses all operations while creating the snapshot: simple but impacts availability. +Copy-on-write uses data structures that support efficient point-in-time copies. Database +transactions can create consistent snapshots if your state is in a database. Some storage +engines support checkpointing to leverage native snapshot capabilities. + + Snapshot Installation and Recovery + +When a peer needs to catch up using a snapshot, it receives the snapshot data from the leader or +loads it from local storage. The state machine is paused to prevent conflicts during restoration, +the snapshot data is loaded replacing any existing state, and the state machine resumes normal +operation by replaying any log entries that occurred after the snapshot. + +Your state machine's `reinitialize` method is responsible for loading snapshots during startup by +loading the latest snapshot if available, with the Raft layer replaying any log entries after +the snapshot. + + Designing Snapshot-Friendly State Machines + +When designing your state machine, ensure your state can be efficiently serialized and +deserialized, avoiding complex object graphs that are difficult to serialize. For very large +state machines, consider whether you can implement incremental snapshots that only capture +changes since the last snapshot. + +If your state machine maintains state in external systems, ensure your snapshot process captures +this external state consistently. Regularly test your snapshot and recovery process to ensure it +works correctly under various failure scenarios. + +### Leadership and Fault Tolerance + +Leadership in Ratis is both simpler and more complex than it might initially appear. Ratis +handles all the mechanics of leader election and failover automatically, but your application +needs to handle leadership changes robustly. Review Comment: Why "your application needs to handle leadership changes robustly" ? Only the applications want to know who is the leader need to handle leadership changes. If an application does not care about who is the leader, then it does not need to do anything. The first sentence is a kind of a personal comment. Let's remove it. I suggest to rewrite this paragraph as below: > Ratis handles all the mechanics of leader election and failover automatically. > If your application does not care about who is the leader, then it does not need to do anything. > Otherwise, your application can optionally observe leadership change and react accordingly; > see [State Machine Leadership Events](#State-Machine-Leadership-Events). ## ratis-docs/src/site/markdown/concept/operations.md: ## @@ -0,0 +1,135 @@ + +# Introduction to
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on PR #1338: URL: https://github.com/apache/ratis/pull/1338#issuecomment-3825331603 Thank you again. :-) I restructured the read consistency section along the lines you suggested and also added a mention and forward reference to the discussion about blocking and async APIs. Thanks! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on PR #1338:
URL: https://github.com/apache/ratis/pull/1338#issuecomment-3822740055
> ... I struggled a bit with reworking the section on read consistency. If
you have any further thoughts on how to make that section clearer or improve
its flow, please let me know.
I have the following suggestion for the read section:
```md
Read Consistency Options
Ratis provides several read patterns with different consistency and
performance characteristics.
Read requests query the state machine of a server directly without going
through the Raft consensus protocol.
The `sendReadOnly()` API sends the request to the leader.
(When a non-leader server receives such request, it throws a
`NotLeaderException`
and then the client will retry other servers.)
In contrast, the `sendReadOnly(message, serverId)` API sends the request to
a particular server,
which may be a leader or a follower.
The server's `raft.server.read.option` configuration affects read behavior:
* **DEFAULT (default setting)**: `sendReadOnly()` performs leader reads for
efficiency.
It provides strong consistency under normal conditions.
* Split-brain Problem: In case that an old leader has been partitioned
from the majority
and a new leader has been elected, reading from the old leader can return
stale data
since the old leader does not have the new transactions committed by the
new leader.
* **LINEARIZABLE**: both `sendReadOnly()` and `sendReadOnly(message,
serverId)`
use the ReadIndex protocol to provide linearizable consistency, ensuring
you always read the most
up-to-date committed data and won't read stale data as described in the
"Split-brain Problem" above.
* Non-linearizable API: Clients may use `sendReadOnlyNonLinearizable()` to
read from leader's state machine
directly without a linearizable guarantee.
Other than the `sendReadOnly(..)` methods mentioned above,
we have the following read APIs:
**Stale reads with minimum index** let you specify a minimum log index that
the peer must have applied
before serving the read. Call `sendStaleRead()`: if the peer hasn't caught
up to your minimum index,
it will throw a `StaleReadException`.
**Asynchronous reads**:
Ratis supports both blocking reads and asynchronous reads.
All the blocking read methods are supported by asynchronous reads
-- the blocking reads return a reply directly while asynchronous reads
return a future of the reply.
Asynchronous reads additionally support the following APIs
* **Read-after-write consistency** ensures reads reflect the latest
successful write by the same client.
Since write requests go through the Raft consensus protocol but read
requests do not,
a read request $R$ may be completed before a write request $W$
even if $R$ is sent (asynchronously) after $W$.
Therefore, the reply returned by $R$ may not include the result updated by
$W$.
Use `sendReadAfterWrite()` when you need to read your own writes immediately.
* **Unorder reads** let the asynchronous read requests complete in any order.
The fastest reply is completed first.
Use `sendReadOnlyUnordered()` when you don't care about the ordering.
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on PR #1338: URL: https://github.com/apache/ratis/pull/1338#issuecomment-3814276924 Good idea about breaking it apart: thanks. I've broken it into five sections/files with navigation between and within sections. You've reviewed the content in sections 1, 2 and 4 so far. The discussion on read consistency is now in section 2. Sections 3 and 5 probably haven't been looked at. It feels like a largish number of sections, but it seems to flow pretty well and offers lots of room for future expansion. Let me know what you think. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on PR #1338: URL: https://github.com/apache/ratis/pull/1338#issuecomment-3813049407 @jcshepherd , thanks for the update! > ... I know it's a longish document. ... How about breaking it to several md files? We can merge the files separately. It would also be easier to read. > ... I struggled a bit with reworking the section on read consistency. If you have any further thoughts on how to make that section clearer or improve its flow, please let me know. Sure, will continue reviewing it tomorrow. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on PR #1338: URL: https://github.com/apache/ratis/pull/1338#issuecomment-3808047669 @szetszwo - Thanks very much for taking the time to review: I know it's a longish document. I've addressed your feedback. I struggled a bit with reworking the section on read consistency. If you have any further thoughts on how to make that section clearer or improve its flow, please let me know. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2734175536 ## ratis-docs/src/site/markdown/concept/index-v2.md: ## @@ -0,0 +1,499 @@ + +# Apache Ratis Concepts + +## Table of Contents + +1. [Overview of Raft and Apache Ratis](#overview-of-raft-and-apache-ratis) +2. [Raft Cluster Topology](#raft-cluster-topology) +3. [The Raft Log - Foundation of Consensus](#the-raft-log---foundation-of-consensus) +4. [The State Machine - Your Application's Heart](#the-state-machine---your-applications-heart) +5. [Consistency Models and Read Patterns](#consistency-models-and-read-patterns) +6. [Snapshots - Managing Growth and Recovery](#snapshots---managing-growth-and-recovery) +7. [Logical Organization of Ratis](#logical-organization-of-ratis) +8. [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) +9. [Scaling with Multi-Raft Groups](#scaling-with-multi-raft-groups) + +## Overview of Raft and Apache Ratis + +The Raft consensus algorithm solves a fundamental problem in distributed systems: how do you get +multiple computers to agree on a sequence of operations, even when some might fail or become +unreachable? This problem, known as distributed consensus, is at the heart of building reliable +distributed systems. + +Raft ensures that a cluster of servers maintains an identical, ordered log of operations. Each +server applies these operations to its local state machine in the same order, guaranteeing that +all servers end up with identical state. This approach, called state machine replication, +provides both consistency and fault tolerance. + +You should consider using Raft when your system needs strong consistency guarantees across +multiple servers. This typically applies to systems where correctness is more important than +absolute performance, such as distributed databases, configuration management systems, or any +application where split-brain scenarios would be unacceptable. + +Apache Ratis is a Java library that implements the Raft consensus protocol. The key word here +is "library" - Ratis is not a standalone service that you communicate with over the network. +Instead, you embed Ratis directly into your Java application, and it becomes part of your +application's runtime. + +This embedded approach creates tight integration between your application and the consensus +mechanism. Your application and Ratis run in the same JVM, sharing memory and computational +resources. Your application provides the business logic (the "state machine" in Raft terminology), +while Ratis handles the distributed consensus mechanics needed to keep multiple instances of your +application synchronized. + +## Raft Cluster Topology + +Understanding the basic building blocks of a Raft deployment affects both the correctness and +performance of your system. + +### Servers, Clusters, and Groups + +A Raft server (also known as a "peer") is a single running instance of your application with +Ratis embedded. Each server runs your state machine and participates in the consensus protocol. + +A Raft cluster is a physical collection of servers that can participate in consensus. A Raft +group is a logical consensus domain that runs across a specific subset of peers in the cluster. +At any given time, one peer in a group acts as the "leader" while the others are "followers" or +"listeners". The leader handles all write requests and replicates operations to other peers in +the group. Both leaders and followers can service read requests, with different consistency +guarantees. + +A single cluster can host multiple independent Raft groups, each with its own leader election, +consistency and state replication. Groups typically consist of an odd number of peers (3, 5, or +7 are common) to ensure clear majority decisions. + +### Majority-Based Decision-Making + +Raft's safety guarantees depend on majority agreement within each group. The leader replicates +each operation to the followers in its group, and operations are committed when at least +(N/2 + 1) peers in that group acknowledge them. This means a group of 3 peers can tolerate 1 +failure, a group of five peers can tolerate 2 failures, and so on. + +This majority requirement affects both availability and performance. A group remains available as +long as a majority of its peers are reachable and functioning. However, every transaction must +wait for majority acknowledgment, so the slowest server in the majority determines your write +latency. + +### Server Placement and Network Considerations + +The physical and network placement of your servers impacts both availability and performance. +Placing all servers in the same rack or data center provides the lowest latency but risks +creating a single point of failure. Distributing servers across multiple availability zones or +data centers improves fault tolerance but can increase latency. + +A common approach is to place servers acr
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2733429957 ## ratis-docs/src/site/markdown/concept/index-v2.md: ## @@ -0,0 +1,499 @@ + +# Apache Ratis Concepts + +## Table of Contents + +1. [Overview of Raft and Apache Ratis](#overview-of-raft-and-apache-ratis) +2. [Raft Cluster Topology](#raft-cluster-topology) +3. [The Raft Log - Foundation of Consensus](#the-raft-log---foundation-of-consensus) +4. [The State Machine - Your Application's Heart](#the-state-machine---your-applications-heart) +5. [Consistency Models and Read Patterns](#consistency-models-and-read-patterns) +6. [Snapshots - Managing Growth and Recovery](#snapshots---managing-growth-and-recovery) +7. [Logical Organization of Ratis](#logical-organization-of-ratis) +8. [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) +9. [Scaling with Multi-Raft Groups](#scaling-with-multi-raft-groups) + +## Overview of Raft and Apache Ratis + +The Raft consensus algorithm solves a fundamental problem in distributed systems: how do you get +multiple computers to agree on a sequence of operations, even when some might fail or become +unreachable? This problem, known as distributed consensus, is at the heart of building reliable +distributed systems. + +Raft ensures that a cluster of servers maintains an identical, ordered log of operations. Each +server applies these operations to its local state machine in the same order, guaranteeing that +all servers end up with identical state. This approach, called state machine replication, +provides both consistency and fault tolerance. + +You should consider using Raft when your system needs strong consistency guarantees across +multiple servers. This typically applies to systems where correctness is more important than +absolute performance, such as distributed databases, configuration management systems, or any +application where split-brain scenarios would be unacceptable. + +Apache Ratis is a Java library that implements the Raft consensus protocol. The key word here +is "library" - Ratis is not a standalone service that you communicate with over the network. +Instead, you embed Ratis directly into your Java application, and it becomes part of your +application's runtime. + +This embedded approach creates tight integration between your application and the consensus +mechanism. Your application and Ratis run in the same JVM, sharing memory and computational +resources. Your application provides the business logic (the "state machine" in Raft terminology), +while Ratis handles the distributed consensus mechanics needed to keep multiple instances of your +application synchronized. + +## Raft Cluster Topology + +Understanding the basic building blocks of a Raft deployment affects both the correctness and +performance of your system. + +### Servers, Clusters, and Groups + +A Raft server (also known as a "peer") is a single running instance of your application with +Ratis embedded. Each server runs your state machine and participates in the consensus protocol. + +A Raft cluster is a physical collection of servers that can participate in consensus. A Raft +group is a logical consensus domain that runs across a specific subset of peers in the cluster. +At any given time, one peer in a group acts as the "leader" while the others are "followers" or +"listeners". The leader handles all write requests and replicates operations to other peers in +the group. Both leaders and followers can service read requests, with different consistency +guarantees. + +A single cluster can host multiple independent Raft groups, each with its own leader election, +consistency and state replication. Groups typically consist of an odd number of peers (3, 5, or +7 are common) to ensure clear majority decisions. + +### Majority-Based Decision-Making + +Raft's safety guarantees depend on majority agreement within each group. The leader replicates +each operation to the followers in its group, and operations are committed when at least +(N/2 + 1) peers in that group acknowledge them. This means a group of 3 peers can tolerate 1 +failure, a group of five peers can tolerate 2 failures, and so on. + +This majority requirement affects both availability and performance. A group remains available as +long as a majority of its peers are reachable and functioning. However, every transaction must +wait for majority acknowledgment, so the slowest server in the majority determines your write +latency. + +### Server Placement and Network Considerations + +The physical and network placement of your servers impacts both availability and performance. +Placing all servers in the same rack or data center provides the lowest latency but risks +creating a single point of failure. Distributing servers across multiple availability zones or +data centers improves fault tolerance but can increase latency. + +A common approach is to place servers acros
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
jcshepherd commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2733376732 ## ratis-docs/src/site/markdown/concept/index-v2.md: ## @@ -0,0 +1,499 @@ + +# Apache Ratis Concepts + +## Table of Contents + +1. [Overview of Raft and Apache Ratis](#overview-of-raft-and-apache-ratis) +2. [Raft Cluster Topology](#raft-cluster-topology) +3. [The Raft Log - Foundation of Consensus](#the-raft-log---foundation-of-consensus) +4. [The State Machine - Your Application's Heart](#the-state-machine---your-applications-heart) +5. [Consistency Models and Read Patterns](#consistency-models-and-read-patterns) +6. [Snapshots - Managing Growth and Recovery](#snapshots---managing-growth-and-recovery) +7. [Logical Organization of Ratis](#logical-organization-of-ratis) +8. [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) +9. [Scaling with Multi-Raft Groups](#scaling-with-multi-raft-groups) + +## Overview of Raft and Apache Ratis + +The Raft consensus algorithm solves a fundamental problem in distributed systems: how do you get +multiple computers to agree on a sequence of operations, even when some might fail or become +unreachable? This problem, known as distributed consensus, is at the heart of building reliable +distributed systems. + +Raft ensures that a cluster of servers maintains an identical, ordered log of operations. Each +server applies these operations to its local state machine in the same order, guaranteeing that +all servers end up with identical state. This approach, called state machine replication, +provides both consistency and fault tolerance. + +You should consider using Raft when your system needs strong consistency guarantees across +multiple servers. This typically applies to systems where correctness is more important than +absolute performance, such as distributed databases, configuration management systems, or any +application where split-brain scenarios would be unacceptable. + +Apache Ratis is a Java library that implements the Raft consensus protocol. The key word here +is "library" - Ratis is not a standalone service that you communicate with over the network. +Instead, you embed Ratis directly into your Java application, and it becomes part of your +application's runtime. + +This embedded approach creates tight integration between your application and the consensus +mechanism. Your application and Ratis run in the same JVM, sharing memory and computational +resources. Your application provides the business logic (the "state machine" in Raft terminology), +while Ratis handles the distributed consensus mechanics needed to keep multiple instances of your +application synchronized. + +## Raft Cluster Topology + +Understanding the basic building blocks of a Raft deployment affects both the correctness and +performance of your system. + +### Servers, Clusters, and Groups + +A Raft server (also known as a "peer") is a single running instance of your application with +Ratis embedded. Each server runs your state machine and participates in the consensus protocol. + +A Raft cluster is a physical collection of servers that can participate in consensus. A Raft +group is a logical consensus domain that runs across a specific subset of peers in the cluster. +At any given time, one peer in a group acts as the "leader" while the others are "followers" or +"listeners". The leader handles all write requests and replicates operations to other peers in +the group. Both leaders and followers can service read requests, with different consistency +guarantees. + +A single cluster can host multiple independent Raft groups, each with its own leader election, +consistency and state replication. Groups typically consist of an odd number of peers (3, 5, or +7 are common) to ensure clear majority decisions. + +### Majority-Based Decision-Making + +Raft's safety guarantees depend on majority agreement within each group. The leader replicates +each operation to the followers in its group, and operations are committed when at least +(N/2 + 1) peers in that group acknowledge them. This means a group of 3 peers can tolerate 1 +failure, a group of five peers can tolerate 2 failures, and so on. + +This majority requirement affects both availability and performance. A group remains available as +long as a majority of its peers are reachable and functioning. However, every transaction must +wait for majority acknowledgment, so the slowest server in the majority determines your write +latency. + +### Server Placement and Network Considerations + +The physical and network placement of your servers impacts both availability and performance. +Placing all servers in the same rack or data center provides the lowest latency but risks +creating a single point of failure. Distributing servers across multiple availability zones or +data centers improves fault tolerance but can increase latency. + +A common approach is to place servers acr
Re: [PR] RATIS-2388 (Further) Enhancing content for concept in ratis-docs [ratis]
szetszwo commented on code in PR #1338: URL: https://github.com/apache/ratis/pull/1338#discussion_r2729645047 ## ratis-docs/src/site/markdown/concept/index-v2.md: ## @@ -0,0 +1,499 @@ + +# Apache Ratis Concepts + +## Table of Contents + +1. [Overview of Raft and Apache Ratis](#overview-of-raft-and-apache-ratis) +2. [Raft Cluster Topology](#raft-cluster-topology) +3. [The Raft Log - Foundation of Consensus](#the-raft-log---foundation-of-consensus) +4. [The State Machine - Your Application's Heart](#the-state-machine---your-applications-heart) +5. [Consistency Models and Read Patterns](#consistency-models-and-read-patterns) +6. [Snapshots - Managing Growth and Recovery](#snapshots---managing-growth-and-recovery) +7. [Logical Organization of Ratis](#logical-organization-of-ratis) +8. [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) +9. [Scaling with Multi-Raft Groups](#scaling-with-multi-raft-groups) + +## Overview of Raft and Apache Ratis + +The Raft consensus algorithm solves a fundamental problem in distributed systems: how do you get +multiple computers to agree on a sequence of operations, even when some might fail or become +unreachable? This problem, known as distributed consensus, is at the heart of building reliable +distributed systems. + +Raft ensures that a cluster of servers maintains an identical, ordered log of operations. Each +server applies these operations to its local state machine in the same order, guaranteeing that +all servers end up with identical state. This approach, called state machine replication, +provides both consistency and fault tolerance. + +You should consider using Raft when your system needs strong consistency guarantees across +multiple servers. This typically applies to systems where correctness is more important than +absolute performance, such as distributed databases, configuration management systems, or any +application where split-brain scenarios would be unacceptable. + +Apache Ratis is a Java library that implements the Raft consensus protocol. The key word here +is "library" - Ratis is not a standalone service that you communicate with over the network. +Instead, you embed Ratis directly into your Java application, and it becomes part of your +application's runtime. + +This embedded approach creates tight integration between your application and the consensus +mechanism. Your application and Ratis run in the same JVM, sharing memory and computational +resources. Your application provides the business logic (the "state machine" in Raft terminology), +while Ratis handles the distributed consensus mechanics needed to keep multiple instances of your +application synchronized. + +## Raft Cluster Topology + +Understanding the basic building blocks of a Raft deployment affects both the correctness and +performance of your system. + +### Servers, Clusters, and Groups + +A Raft server (also known as a "peer") is a single running instance of your application with +Ratis embedded. Each server runs your state machine and participates in the consensus protocol. + +A Raft cluster is a physical collection of servers that can participate in consensus. A Raft +group is a logical consensus domain that runs across a specific subset of peers in the cluster. +At any given time, one peer in a group acts as the "leader" while the others are "followers" or Review Comment: Since a group could temporarily have no leader or more than one leaders (old leader not yet timed out), let remove "At any given time, ", i.e. > One of the peers in a group acts as the "leader" ... ## ratis-docs/src/site/markdown/concept/index-v2.md: ## @@ -0,0 +1,499 @@ + +# Apache Ratis Concepts + +## Table of Contents + +1. [Overview of Raft and Apache Ratis](#overview-of-raft-and-apache-ratis) +2. [Raft Cluster Topology](#raft-cluster-topology) +3. [The Raft Log - Foundation of Consensus](#the-raft-log---foundation-of-consensus) +4. [The State Machine - Your Application's Heart](#the-state-machine---your-applications-heart) +5. [Consistency Models and Read Patterns](#consistency-models-and-read-patterns) +6. [Snapshots - Managing Growth and Recovery](#snapshots---managing-growth-and-recovery) +7. [Logical Organization of Ratis](#logical-organization-of-ratis) +8. [Leadership and Fault Tolerance](#leadership-and-fault-tolerance) +9. [Scaling with Multi-Raft Groups](#scaling-with-multi-raft-groups) + +## Overview of Raft and Apache Ratis + +The Raft consensus algorithm solves a fundamental problem in distributed systems: how do you get +multiple computers to agree on a sequence of operations, even when some might fail or become +unreachable? This problem, known as distributed consensus, is at the heart of building reliable +distributed systems. + +Raft ensures that a cluster of servers maintains an identical, ordered log of operations. Each +server applies these operations to its local state machine in the sam
