[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17295880#comment-17295880 ] Sagar Rao commented on KAFKA-10526: --- hey [~hachikuji] sorry for bugging again on this but could you plz help me out with the queries above? > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Assignee: Sagar Rao >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17291485#comment-17291485 ] Sagar Rao commented on KAFKA-10526: --- [~hachikuji], I looked at the codebase and the KIP further and here's what I understood: 1) Any new records that the leader receives, it immediately updates its local state. This happens via the maybeAppendBatches method which invokes flushLeaderLog. In flushLeaderLog, for the bunch of records, it would update it's local state and check if the HWM can be advanced. Note that after this step, the log is always flushed to disk in flushLeaderLog. 2) The followers invoke fetch requests to fetch records. Once the leader receives such a message, it invokes tryCompleteFetchRequest which validates the request. At this point, it reads a bunch of records which can be returned to the follower and it tries to update the replicaState. It also tries to update the HWM and if it does, then the HWM on the log is also advanced. 3) The follower, when it receives a FetchResponse, appends the response to its log and also flushes the record to its log. I believe it also updates the follower watermark here. So, in this flow, flush happens in 2 flows: 1) when the leader completes a batch and secondly, when a fetchresponse is received by the follower. As per the Op in the ticket, fsync is called a number of times on the followers, so that is the ls the latter. Few questions that I have: 1) Basic question, but I see all this logic in KafkaRaftClient. where does the instance of the class get instantiated? Is it on the leader? 2) looking at this flow, i am slightly confused on how does the leader know which records have been committed successfully on the followers? It seems to maintian a local copy of replicas and their offsets and epochs, but how does it know which have been committed? Is it via the fetch requests received from the followers? 3) The optimisation that you have talked about, where does that need to happen in this flow? Is it while handling fetch responses or when appending new records in the batch? Or is it some other place? > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Assignee: Sagar Rao >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17291436#comment-17291436 ] Sagar Rao commented on KAFKA-10526: --- hi [~hachikuji], can you plz validate my understanding from the above comment whenever you get the chance? > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Assignee: Sagar Rao >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17282971#comment-17282971 ] Sagar Rao commented on KAFKA-10526: --- [~hachikuji], I have looked at the codebase and also the KIP-595 and tried to understand this. One thing that I want to know is that the log replication happens via the Fetch request/response dance. So, the leader gets a Fetch request and if all pre conditions are met, finds a bunch of records and returns a FetchResponse. During that process, it keeps updating its LocalState and the replicated state for each Fetch Request that comes through. In that process, it also tries to check if the highwatermark can be moved ahead as it tries to find if a majority of followers are at a point > current HMW offset. The follower, when it receives the FetchResponse, looks at the messages and see if it needs to truncate its log or if the LEader has been fenced etc and then finally, writes the records passed in the FetchResponse to its log. What I am not able to figure out is that how does the leader know that a write has been committed on the follower side. I could find the code to check if the HWM should be incremented or not based upon the ReplicaState. The FetchResponse handler finally returns if the fetch was successful or not, but how is the value propagated back to the leader? There are some listener contexts, is it through that or via the NetworkChannels? I see a correlation id which is being used in the Raftinbound messages as well. In terms of the optimisation that you have suggested, instead of updating LocalState/ReplicaState every time as the leader receives each FetchRequest it can wait if the majority has committed the writes and flush only then. Is that the correct understanding? > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Assignee: Sagar Rao >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17262724#comment-17262724 ] Sagar Rao commented on KAFKA-10526: --- [~hachikuji], while KAFKA-10652 is getting reviewed, I was wondering if I can get started on this one. Any pointers/docs on how to perform the benchmarking? I will also start looking at the points where fsync deferral can be made post the crossing of high watermark. > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Assignee: Sagar Rao >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=1746#comment-1746 ] Sagar Rao commented on KAFKA-10526: --- Sure thank you [~hachikuji]!, i have assigned this one and the other 2 to myself. I will go through the code > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17221881#comment-17221881 ] Jason Gustafson commented on KAFKA-10526: - [~sagarrao] Yes, of course. I might suggest KAFKA-10652 as a lower hanging fruit to get into the code a little bit. > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (KAFKA-10526) Explore performance impact of leader fsync deferral
[ https://issues.apache.org/jira/browse/KAFKA-10526?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17215353#comment-17215353 ] Sagar Rao commented on KAFKA-10526: --- hey [~hachikuji], is it something that I can pick up? I am not sure if you or someone else is planning to pick it as it seems to be related to the raft protocol KIP.. Let me know plz. > Explore performance impact of leader fsync deferral > --- > > Key: KAFKA-10526 > URL: https://issues.apache.org/jira/browse/KAFKA-10526 > Project: Kafka > Issue Type: Sub-task >Reporter: Jason Gustafson >Priority: Major > > In order to commit a write, a majority of nodes must call fsync in order to > ensure the data has been written to disk. An interesting optimization option > to consider is letting the leader defer fsync until the high watermark is > ready to be advanced. This potentially allows us to reduce the number of > flushes on the leader. -- This message was sent by Atlassian Jira (v8.3.4#803005)