From 3f5ab3f9824cd128d0029a2402fca53fbbbf48a8 Mon Sep 17 00:00:00 2001 From: Sander Saares Date: Fri, 22 Oct 2021 09:02:42 +0300 Subject: [PATCH] A CPU request does not result in a guarantee It is merely a pod placement constraint but the phrasing leads readers to the incorrect understanding that some actual CPU time is guaranteed/reserved. Re-phrase to make it clear this is nothing more than a request. --- .../concepts/configuration/manage-resources-containers.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/content/en/docs/concepts/configuration/manage-resources-containers.md b/content/en/docs/concepts/configuration/manage-resources-containers.md index cbcfff67ef..3f0d3f9da6 100644 --- a/content/en/docs/concepts/configuration/manage-resources-containers.md +++ b/content/en/docs/concepts/configuration/manage-resources-containers.md @@ -100,7 +100,7 @@ Limits and requests for CPU resources are measured in *cpu* units. One cpu, in Kubernetes, is equivalent to **1 vCPU/Core** for cloud providers and **1 hyperthread** on bare-metal Intel processors. Fractional requests are allowed. A Container with -`spec.containers[].resources.requests.cpu` of `0.5` is guaranteed half as much +`spec.containers[].resources.requests.cpu` of `0.5` asks for half as much CPU as one that asks for 1 CPU. The expression `0.1` is equivalent to the expression `100m`, which can be read as "one hundred millicpu". Some people say "one hundred millicores", and this is understood to mean the same thing. A @@ -236,7 +236,7 @@ The kubelet also uses this kind of storage to hold container images, and the writable layers of running containers. {{< caution >}} -If a node fails, the data in its ephemeral storage can be lost. +If a node fails, the data in its ephemeral storage can be lost. Your applications cannot expect any performance SLAs (disk IOPS for example) from local ephemeral storage. {{< /caution >}} @@ -440,7 +440,7 @@ Kubernetes does not use them. Quotas are faster and more accurate than directory scanning. When a directory is assigned to a project, all files created under a directory are created in that project, and the kernel merely has to -keep track of how many blocks are in use by files in that project. +keep track of how many blocks are in use by files in that project. If a file is created and deleted, but has an open file descriptor, it continues to consume space. Quota tracking records that space accurately whereas directory scans overlook the storage used by deleted files.