[zh] Update reserve-compute-resources.md

Signed-off-by: xin.li <xin.li@daocloud.io>

content/zh/docs/tasks/administer-cluster/reserve-compute-resources.md

@@ -166,7 +166,7 @@ flag.
 It is recommended that the kubernetes system daemons are placed under a top
 level control group (`runtime.slice` on systemd machines for example). Each
 system daemon should ideally run within its own child control group. Refer to
-[the design proposal](https://git.k8s.io/community/contributors/design-proposals/node/node-allocatable.md#recommended-cgroups-setup)
+[the design proposal](https://git.k8s.io/design-proposals-archive/node/node-allocatable.md#recommended-cgroups-setup)
 for more details on recommended control group hierarchy.
 
 Note that Kubelet **does not** create `--kube-reserved-cgroup` if it doesn't
@@ -179,7 +179,7 @@ exist. Kubelet will fail if an invalid cgroup is specified.
 `runtime.slice`）。
 理想情况下每个系统守护进程都应该在其自己的子控制组中运行。
 请参考
-[这个设计方案](https://git.k8s.io/community/contributors/design-proposals/node/node-allocatable.md#recommended-cgroups-setup)，
+[这个设计方案](https://git.k8s.io/design-proposals-archive/node/node-allocatable.md#recommended-cgroups-setup)，
 进一步了解关于推荐控制组层次结构的细节。
 
 请注意，如果 `--kube-reserved-cgroup` 不存在，Kubelet 将 **不会** 创建它。