Status

Current state: Under DiscussionAccepted

Discussion thread: here

JIRA: KAFKA-2063

Released: 0.10.1.0

Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).

...

Currently the only possible way for client to limit fetch response size is via per-partition response limit max_bytes taken from config setting max.partition.fetch.bytes.

So the maximum amount of memory the client can consume is max.partition.fetch.bytes * num_partitions, where num_partitions is the total number of partitions currently being fetched by consumer.

This leads to following problems:

1. Since num_partitions can be quite big (several thousands), the memory required for fetch response responses can be several GB
2. max.partition.fetch.bytes can not be set arbitrarily low since it should be greater than maximum message size for fetch request to work.
3. Memory usage is not easily predictable - it depends on consumer lag

This KIP proposes to introduce new version of fetch request with new top-level parameter "response_max_bytes" to limit the size of fetch response and solve above problem.

In particular, if consumer issues N parallel fetch requests, the memory consumption will not exceed N * max_bytes.

Actually, it will be min(N * max_bytes,  max.partition.fetch.bytes * num_partitions)  since per-partition limit is still respected.

Public Interfaces

This KIP introduces:

• New fetch request (v.3) with response size limit
• New client-side config parameter fetch.response.max.bytes - client's fetch response size limit
• New replication config parameter replica.fetch.response.max.bytes - limit used by replication thread
• New inter-broker protocol version "0.1110.01-IV0" - starting from this version brokers will use fetch request v.3 for replication 

...

Proposed changes are quite straightforward. We introduce FetchRequest v.3 with new top level parameter max_bytes:

Fetch Request (Version: 3) => replica_id max_wait_time min_bytes max_bytes [topics]
  replica_id => INT32 
  max_wait_time => INT32
  min_bytes => INT32
  max_bytes => INT32
  topics => topic [partitions]
    topic => STRING
    partitions => partition fetch_offset max_bytes
      partition => INT32
      fetch_offset => INT64
      max_bytes => INT32

Fetch Response v.3 will remain the same as v.2.

Server . New fetch request processes partitions in order they appear in request.

If response_max_bytes parameter is zero ("no limit"), the request is processed exactly as before.

Otherwise, for each partition except the first one server fetches up to corresponding partition limit parition_max_bytes, but not bigger than remaining response limit.

For the first partition, server always fetches at least least one message. max.bytes. Empty response limits will be returned for all partitions that didn't fit into response limit.

This algorithm provides following guarantees:

• FetchRequest with non-zero response_max_bytes always makes progress - if server has message(s), than at least one message is returned irrespective of response_max_bytes  parameter
• FetchRequest response size will not be bigger than max(response_max_bytes, message.max.bytes)

...

• size of the first message in first partition)

Since new fetch request processes partitions in order and stops fetching data when response limit is hit, client should use some kind of partition shuffling to ensure fairness.

...

In this scenario client won't  get any messages from C and D until it catches up with A and B.

The solution is to reorder partitions in fetch request in round-robin fashion to continue fetching from first empty partition in round-robin fashion received or to perform random shuffle of partitions before each request.

Round-robin shuffling seems to be more "fair" and predictable so we decided to deploy it at ReplicaFetcherThread and in Consumer Java API.

...

Compatibility, Deprecation, and Migration Plan

New The new fetch request is designed to work properly even if response_the top level max_bytes is less than the message size. If response_max_bytes is zero, new requests behaves exactly like old one.

So we can even make this KIP absolutely transparent for users by making setting default for both fetch.response.max.bytes and replica.fetch.response.max.bytes to zero.

However, since clients like ReplicaFetcherThread and Java Consumer are ready for new fetch request, we decided to enable   We decided to establish the following defaults:

fetch.response.max.bytes = 50MB

replica.fetch.response.max.bytes = 10MB

...

1. Together with addition of global response limit deprecate per-partitions limit. Rejected since per-partition limit can be useful for Kafka streams (see mail list discussion). 
2. Do random partition shuffling on server side. Pros: ensure fairness without client-side modifications. Cons: non-deterministic behaviour on server side; round-robin can be easily implemented on client side.