[
https://issues.apache.org/jira/browse/HDDS-14255?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ivan Andika updated HDDS-14255:
-------------------------------
Description:
We need to specify the consistency guarantee for OM HA, SCM HA, DN (BCSID).
For example
* The default OM HA consistency (lease and linearizable read disabled) does
not require heartbeat rounds for each write or read requests, but might have a
short-period of split brain (if there are one new leader and old leader).
* Unlike OM and SCM HA, DN ContainerStateMachine allows multiple
StateMachine#applyTransaction to executed at one time (
hdds.container.ratis.num.container.op.executors)to maximize throughput.
* For DN, we also need to update the durability guarantee similar to
https://hadoop.apache.org/docs/stable/hadoop-project-dist/hadoop-common/filesystem/outputstream.html#How_data_is_written_to_a_filesystem
was:
We need to specify the consistency guarantee for OM HA, SCM HA, DN (BCSID).
For example
* The default OM HA consistency (lease and linearizable read disabled) does
not require heartbeat rounds for each write or read requests, but might have a
short-period of split brain (if there are one new leader and old leader).
* Unlike OM and SCM HA, DN ContainerStateMachine allows multiple
StateMachine#applyTransaction to executed at one time (
hdds.container.ratis.num.container.op.executors)to maximize throughput.
> [Website v2] [Docs] [Core Concepts] Consistency Guarantee
> ---------------------------------------------------------
>
> Key: HDDS-14255
> URL: https://issues.apache.org/jira/browse/HDDS-14255
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: Ivan Andika
> Assignee: Chu Cheng Li
> Priority: Major
> Labels: pull-request-available
>
> We need to specify the consistency guarantee for OM HA, SCM HA, DN (BCSID).
> For example
> * The default OM HA consistency (lease and linearizable read disabled) does
> not require heartbeat rounds for each write or read requests, but might have
> a short-period of split brain (if there are one new leader and old leader).
> * Unlike OM and SCM HA, DN ContainerStateMachine allows multiple
> StateMachine#applyTransaction to executed at one time (
> hdds.container.ratis.num.container.op.executors)to maximize throughput.
> * For DN, we also need to update the durability guarantee similar to
> https://hadoop.apache.org/docs/stable/hadoop-project-dist/hadoop-common/filesystem/outputstream.html#How_data_is_written_to_a_filesystem
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
