-
Notifications
You must be signed in to change notification settings - Fork 593
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ducktape: Increase partition density to 2k per shard in the MPT #24710
base: dev
Are you sure you want to change the base?
Conversation
Increase partition density to 2k partitions per shard and adapt the test to use the new memory group aware config.
kgo-repeaters start to get pretty loaded at higher partition densities so the status query can take a bit longer.
478891d
to
417bb75
Compare
Partial fix for https://redpandadata.atlassian.net/browse/CORE-8732. |
@@ -43,12 +43,16 @@ | |||
BIG_FETCH = 104857600 | |||
|
|||
# How much memory to assign to redpanda per partition. Redpanda will be started | |||
# with MIB_PER_PARTITION * PARTITIONS_PER_SHARD * CORE_COUNT memory | |||
DEFAULT_MIB_PER_PARTITION = 4 | |||
# with (MIB_PER_PARTITION * PARTITIONS_PER_SHARD * CORE_COUNT) / (PARTITIONS_MEMORY_ALLOCATION_PERCENT / 100) memory |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should be / (PARTITIONS_MEMORY_ALLOCATION_PERCENT
be * / (PARTITIONS_MEMORY_ALLOCATION_PERCENT)
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
* / (PARTITIONS_MEMORY_ALLOCATION_PERCENT)
Do you mean * 100 / PARTITIONS_MEMORY_ALLOCATION_PERCENT
?
|
||
# How many partitions we will create per shard: this is the primary scaling | ||
# factor that controls how many partitions a given cluster will get. | ||
DEFAULT_PARTITIONS_PER_SHARD = 1000 | ||
DEFAULT_PARTITIONS_PER_SHARD = 2000 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Almost feels like we should read it from config, but anyway LGTM.
partition_replicas_per_node + per_node_slack) | ||
memory_setting = mib_per_partition * ( | ||
partition_replicas_per_node + | ||
per_node_slack) * (self.partition_memory_reserve_percentage / 100.) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Miscommitted some last minute changes I was playing around with. Needs to be /
Increase the partition density in the MPT to 2k per shard and use the new memory group aware config.
As discussed the plan is to raise this to 3k but that makes us currently run into an oversized alloc (large enough to fail the test) so that needs to be fixed first before we can increase the density in the test further (see https://redpandadata.slack.com/archives/C04B5T58QQ1/p1736265968267079).
Backports Required
Release Notes