...

1. batch.size: it specifies a maximum batch size in bytes per partition (default 16384)
2. linger.ms: it specifies a maximum duration to fill the batch in milliseconds (default 0 or no delay)

...

However, when we set the batch size, we'll run into one trade-off situationa dilemma:

1. either we have higher throughput + more memory waste,
2. or we have slower throughput + less memory waste.

...

That explains why we have the trade-off dilemma when setting this config.

...

In this KIP, I'm going to introduce 2 new configuration for users to configa dynamic expandable buffer size for producer.

Public Interfaces

2 Producer config will be introduced:

1. batch.initial.size: It specifies the initial batch size when new batch created (default is the same as 0(disabled), which means we'll always allocate "batch.size" , which means buffer and no buffer expansion will happen)
2. batch.reallocation.factor: it specifies the factor when we try to reallocate the new buffer (default is 2)

Public Interfaces

Briefly list any new interfaces that will be introduced as part of this proposal or any existing interfaces that will be removed or changed. The purpose of this section is to concisely call out the public contract that will come along with this feature.

A public interface is any change to the following:

• Binary log format

• The network protocol and api behavior

• Any class in the public packages under clientsConfiguration, especially client configuration

  • org/apache/kafka/common/serialization

  • org/apache/kafka/common

  • org/apache/kafka/common/errors

  • org/apache/kafka/clients/producer

  • org/apache/kafka/clients/consumer (eventually, once stable)

• Monitoring

• Command line tools and arguments

• Anything else that will likely break existing users in some way when they upgrade

Proposed Changes

To have a better memory usage, the relation of the configurations is recommended to be: "batch.size" = "batch.initial.size" * "batch.reallocation.factor"^n (n means the times we expansion)

ex: "batch.size" = 16KB, "batch.initial.size" = 2KB, "batch.reallocation.factor" = 2

16KB = 2KB * 2^3

Proposed Changes

We'll allocate the batch.initial.size memory when new records send to an empty partition. While we accumulated more records in the partitions to reach the "batch.initial.size" (ex: 2KB), we'll do buffer expansion to the "batch.reallocation.factor" size (ex: 2KB * 2 = 4KB), and keeps accumulating the records, until we reach the "batch.size", or the "linger.ms" expired.

Please note, the buffer expansion is an array copy process (internally we use ByteBuffer), so it's not a free operation. Please also consider the cost of expansion, and set a reasonable "batch.initial.size"Describe the new thing you want to do in appropriate detail. This may be fairly extensive and have large subsections of its own. Or it may be a few sentences. Use judgement based on the scope of the change.

Compatibility, Deprecation, and Migration Plan

• What impact (if any) will there be on existing users?
• If we are changing behavior how will we phase out the older behavior?
• If we need special migration tools, describe them here.
• When will we remove the existing behavior?

Rejected Alternatives

Because the "batch.initial.size" default value is 0(disabled), which means we'll always allocate "batch.size" buffer and no buffer expansion will happen, there will be always backward compatible. No migration plan is needed.

Rejected Alternatives

If there are alternative ways of accomplishing the same thing, what were they? The purpose of this section is to motivate why the design is the way it is and not some other way.n/a