[
https://issues.apache.org/jira/browse/KAFKA-2063?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15401087#comment-15401087
]
Andrey Neporada commented on KAFKA-2063:
----------------------------------------
Thank you all for feedback!
1. I agree that it makes sense to remove partition level maxBytes from fetch
request - having two limits in a single request looks like overkill IMHO.
Server-side max.partition.fetch.bytes should be enough for now.
2. Maybe we should stick with deterministic processing of fetch request on
server side for now. But I would propose to make it possible for server to
reorder partitions to reduce request latency (for example server can take into
account current partition IO load and reorder partitions accordingly). Client
just don't have enough information to come up with most efficient partition
ordering.
So, my proposals are:
a) remove per-partition limit from fetch request v.3, use server-side setting
for this limit
b) (optional) document that starting from fetch request v.3 server _may_
reorder partitions to reduce response time
c) make fetch request deterministic
d) do random shuffle on client side (in ReplicaFetcherThread)
> Bound fetch response size
> -------------------------
>
> Key: KAFKA-2063
> URL: https://issues.apache.org/jira/browse/KAFKA-2063
> Project: Kafka
> Issue Type: Improvement
> Reporter: Jay Kreps
>
> Currently the only bound on the fetch response size is
> max.partition.fetch.bytes * num_partitions. There are two problems:
> 1. First this bound is often large. You may chose
> max.partition.fetch.bytes=1MB to enable messages of up to 1MB. However if you
> also need to consume 1k partitions this means you may receive a 1GB response
> in the worst case!
> 2. The actual memory usage is unpredictable. Partition assignment changes,
> and you only actually get the full fetch amount when you are behind and there
> is a full chunk of data ready. This means an application that seems to work
> fine will suddenly OOM when partitions shift or when the application falls
> behind.
> We need to decouple the fetch response size from the number of partitions.
> The proposal for doing this would be to add a new field to the fetch request,
> max_bytes which would control the maximum data bytes we would include in the
> response.
> The implementation on the server side would grab data from each partition in
> the fetch request until it hit this limit, then send back just the data for
> the partitions that fit in the response. The implementation would need to
> start from a random position in the list of topics included in the fetch
> request to ensure that in a case of backlog we fairly balance between
> partitions (to avoid first giving just the first partition until that is
> exhausted, then the next partition, etc).
> This setting will make the max.partition.fetch.bytes field in the fetch
> request much less useful and we should discuss just getting rid of it.
> I believe this also solves the same thing we were trying to address in
> KAFKA-598. The max_bytes setting now becomes the new limit that would need to
> be compared to max_message size. This can be much larger--e.g. setting a 50MB
> max_bytes setting would be okay, whereas now if you set 50MB you may need to
> allocate 50MB*num_partitions.
> This will require evolving the fetch request protocol version to add the new
> field and we should do a KIP for it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
