Merge pull request kubernetes#6290 from faan11/master

Clarify Scale down rule in the documentation (FAQ.md)

cluster-autoscaler/FAQ.md

@@ -590,7 +590,7 @@ any nodes left unregistered after this time.
 Every 10 seconds (configurable by `--scan-interval` flag), if no scale-up is
 needed, Cluster Autoscaler checks which nodes are unneeded. A node is considered for removal when **all** below conditions hold:
 
-* The sum of cpu and memory requests of all pods running on this node ([DaemonSet pods](https://kubernetes.io/docs/concepts/workloads/controllers/daemonset/) and [Mirror pods](https://kubernetes.io/docs/tasks/configure-pod-container/static-pod/) are included by default but this is configurable with `--ignore-daemonsets-utilization` and `--ignore-mirror-pods-utilization` flags) is smaller
+* The sum of cpu requests and sum of memory requests of all pods running on this node ([DaemonSet pods](https://kubernetes.io/docs/concepts/workloads/controllers/daemonset/) and [Mirror pods](https://kubernetes.io/docs/tasks/configure-pod-container/static-pod/) are included by default but this is configurable with `--ignore-daemonsets-utilization` and `--ignore-mirror-pods-utilization` flags) are smaller
   than 50% of the node's allocatable. (Before 1.1.0, node capacity was used
   instead of allocatable.) Utilization threshold can be configured using
   `--scale-down-utilization-threshold` flag.
@@ -770,7 +770,7 @@ The following startup parameters are supported for cluster autoscaler:
 | `scale-down-delay-after-failure` | How long after scale down failure that scale down evaluation resumes | 3 minutes
 | `scale-down-unneeded-time` | How long a node should be unneeded before it is eligible for scale down | 10 minutes
 | `scale-down-unready-time` | How long an unready node should be unneeded before it is eligible for scale down | 20 minutes
-| `scale-down-utilization-threshold` | Node utilization level, defined as sum of requested resources divided by capacity, below which a node can be considered for scale down | 0.5
+| `scale-down-utilization-threshold` | The maximum value between the sum of cpu requests and sum of memory requests of all pods running on the node divided by node's corresponding allocatable resource, below which a node can be considered for scale down. This value is a floating point number that can range between zero and one. | 0.5
 | `scale-down-non-empty-candidates-count` | Maximum number of non empty nodes considered in one iteration as candidates for scale down with drain<br>Lower value means better CA responsiveness but possible slower scale down latency<br>Higher value can affect CA performance with big clusters (hundreds of nodes)<br>Set to non positive value to turn this heuristic off - CA will not limit the number of nodes it considers." | 30
 | `scale-down-candidates-pool-ratio` | A ratio of nodes that are considered as additional non empty candidates for<br>scale down when some candidates from previous iteration are no longer valid<br>Lower value means better CA responsiveness but possible slower scale down latency<br>Higher value can affect CA performance with big clusters (hundreds of nodes)<br>Set to 1.0 to turn this heuristics off - CA will take all nodes as additional candidates.  | 0.1
 | `scale-down-candidates-pool-min-count` | Minimum number of nodes that are considered as additional non empty candidates<br>for scale down when some candidates from previous iteration are no longer valid.<br>When calculating the pool size for additional candidates we take<br>`max(#nodes * scale-down-candidates-pool-ratio, scale-down-candidates-pool-min-count)` | 50

cluster-autoscaler/main.go

@@ -122,7 +122,7 @@ var (
 	scaleDownUnreadyTime = flag.Duration("scale-down-unready-time", config.DefaultScaleDownUnreadyTime,
 		"How long an unready node should be unneeded before it is eligible for scale down")
 	scaleDownUtilizationThreshold = flag.Float64("scale-down-utilization-threshold", config.DefaultScaleDownUtilizationThreshold,
-		"Sum of cpu or memory of all pods running on the node divided by node's corresponding allocatable resource, below which a node can be considered for scale down")
+		"The maximum value between the sum of cpu requests and sum of memory requests of all pods running on the node divided by node's corresponding allocatable resource, below which a node can be considered for scale down")
 	scaleDownGpuUtilizationThreshold = flag.Float64("scale-down-gpu-utilization-threshold", config.DefaultScaleDownGpuUtilizationThreshold,
 		"Sum of gpu requests of all pods running on the node divided by node's allocatable resource, below which a node can be considered for scale down."+
 			"Utilization calculation only cares about gpu resource for accelerator node. cpu and memory utilization will be ignored.")