From b421bd544b71caa6abcff0b84c3b47569ca876fd Mon Sep 17 00:00:00 2001 From: Erik Sundell Date: Fri, 12 Apr 2024 07:59:43 +0200 Subject: [PATCH] docs: detail inline heading update Co-authored-by: Georgiana --- dashboards/cluster.jsonnet | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/dashboards/cluster.jsonnet b/dashboards/cluster.jsonnet index a6fb2ef..a7d0905 100755 --- a/dashboards/cluster.jsonnet +++ b/dashboards/cluster.jsonnet @@ -91,7 +91,7 @@ local nodepoolMemoryCommitment = ||| % of memory in each node pool guaranteed to user workloads. - Common shapes: + Common shapes this visualization may take: 1. If this is consistently low (<50%), you are paying for cloud compute that you do not need. Consider reducing the size of your nodes, or increasing the amount of memory guaranteed to your users. Some variability based on time of day is to be expected. @@ -150,7 +150,7 @@ local nodepoolCPUCommitment = Most commonly, JupyterHub workloads are *memory bound*, not CPU bound. So this is not a particularly helpful graph. - Common shapes: + Common shapes this visualization may take: 1. If this is *consistently high* but shaped differently than your memory commitment graph, consider changing your CPU requirements. |||