Skip to content
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

chore: add a configuraiton toml for serving-only CN #13239

Merged
merged 4 commits into from
Nov 13, 2023
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 14 additions & 0 deletions src/config/serving-only.toml
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
[storage]
# Shared buffer is useless to a serving-only compute node.
shared_buffer_capacity_mb = 1
# Compactor is irrelevant to a serving-only compute node.
compactor_memory_limit_mb = 1
# Suppose a serving-only CN has 16GB memory in total, we recommend using 30% by default: 16*0.3=4.8
block_cache_capacity_mb = 4800
# Suppose a serving-only CN has 16GB memory in toal, we recommend using 10% by default: 16*0.1=1.6
meta_cache_capacity_mb = 1600
# There will be 16GB - 4.8GB - 1.6GB - reserved memory 16GB * 0.2 (recommended and hard coded by developers) = 6.4GB for executing serving queries. We call it "compute memory".

# If one batch query is resource-intensive and its runtime memory consumption exceeds 6.4GB, it will kill itself automatically before triggering out-of-memory error. The reserved memory is allocated for all types of temporary memory usage and slackness.
# Although we don't recommend users to execute OLAP-style batch queries that take a huge amount of input data, e.g. more than dozens of GB, users can adjust the configuration if such batch query is needed and the default configuration above leaves too few "compute memory".
# Feel free to allocate less memory for "block cache" and "meta cache" to increase "compute memory".
Loading