[
https://issues.apache.org/jira/browse/CASSANDRA-11521?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15231896#comment-15231896
]
Stefania commented on CASSANDRA-11521:
--------------------------------------
bq. I'd like us to tackle this and CASSANDRA-11520 as independently as
possible, since as much as they both are meant to help the same scenario, they
are independent optimisations.
In the proof of concept they are not independent, you can take a look at [this
code|https://github.com/stef1927/cassandra/blob/9259/src/java/org/apache/cassandra/service/BulkReadService.java]
when you get a chance. It should give you a very good idea of what I had in
mind, we basically stream results without even stopping the iteration.
I can see how they can be independent, and I'll be sure to share the design
upfront if that's the path we choose, but wouldn't this mean that internally we
still process each page independently, which means creating sstable iterators
for every single page for example.
Do you think the approach for CL.ONE that I chose in the proof of concept would
be too problematic for resource management?
> Implement streaming for bulk read requests
> ------------------------------------------
>
> Key: CASSANDRA-11521
> URL: https://issues.apache.org/jira/browse/CASSANDRA-11521
> Project: Cassandra
> Issue Type: Sub-task
> Components: Local Write-Read Paths
> Reporter: Stefania
> Assignee: Stefania
> Fix For: 3.x
>
>
> Allow clients to stream data from a C* host, bypassing the coordination layer
> and eliminating the need to query individual pages one by one.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
