Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Kafka >> mail # dev >> Review Request 18299: Proposed In-place Compression on MemoryRecords


Copy link to this message
-
Re: Review Request 18299: Proposed In-place Compression on MemoryRecords

This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/18299/#review34977

clients/src/main/java/org/apache/kafka/common/record/CompressedMemoryRecords.java
<https://reviews.apache.org/r/18299/#comment65371>

    This should maybe be lazy?

clients/src/main/java/org/apache/kafka/common/record/CompressedMemoryRecords.java
<https://reviews.apache.org/r/18299/#comment65370>

    Writing the full record array is not actually correct. The ByteBuffer just represents a slice of the backing array so if you write the full backing array you may be writing other data.

clients/src/main/java/org/apache/kafka/common/record/CompressedMemoryRecords.java
<https://reviews.apache.org/r/18299/#comment65368>

    I don't think this will fly, we should refactor in some way to do the right in place.

clients/src/main/java/org/apache/kafka/common/record/CompressedMemoryRecords.java
<https://reviews.apache.org/r/18299/#comment65369>

    We should ideally do this with out double-copying.
- Jay Kreps
On Feb. 20, 2014, 2:58 a.m., Guozhang Wang wrote: