[autoscaler] Remove legacy autoscaler (#11802)

python/ray/autoscaler/aws/defaults.yaml

@@ -9,28 +9,17 @@ min_workers: 0
 # node. This takes precedence over min_workers.
 max_workers: 2
 
-# The initial number of worker nodes to launch in addition to the head
-# node. When the cluster is first brought up (or when it is refreshed with a
-# subsequent `ray up`) this number of nodes will be started.
-initial_workers: 0
-
-# Whether or not to autoscale aggressively. If this is enabled, if at any point
-#   we would start more workers, we start at least enough to bring us to
-#   initial_workers.
-autoscaling_mode: default
+# The autoscaler will scale up the cluster faster with higher upscaling speed.
+# E.g., if the task requires adding more nodes then autoscaler will gradually
+# scale up the cluster in chunks of upscaling_speed*currently_running_nodes.
+# This number should be > 0.
+upscaling_speed: 1.0
 
 # This executes all commands on all nodes in the docker container,
 # and opens all the necessary ports to support the Ray cluster.
 # Empty string means disabled.
 docker: {}
 
-# The autoscaler will scale up the cluster to this target fraction of resource
-# usage. For example, if a cluster of 10 nodes is 100% busy and
-# target_utilization is 0.8, it would resize the cluster to 13. This fraction
-# can be decreased to increase the aggressiveness of upscaling.
-# This max value allowed is 1.0, which is the most conservative setting.
-target_utilization_fraction: 0.8
-
 # If a node is idle for this many minutes, it will be removed.
 idle_timeout_minutes: 5
 

python/ray/autoscaler/aws/development-example.yaml

@@ -9,12 +9,11 @@ min_workers: 2
 # node. This takes precedence over min_workers.
 max_workers: 2
 
-# The autoscaler will scale up the cluster to this target fraction of resource
-# usage. For example, if a cluster of 10 nodes is 100% busy and
-# target_utilization is 0.8, it would resize the cluster to 13. This fraction
-# can be decreased to increase the aggressiveness of upscaling.
-# This value must be less than 1.0 for scaling to happen.
-target_utilization_fraction: 0.8
+# The autoscaler will scale up the cluster faster with higher upscaling speed.
+# E.g., if the task requires adding more nodes then autoscaler will gradually
+# scale up the cluster in chunks of upscaling_speed*currently_running_nodes.
+# This number should be > 0.
+upscaling_speed: 1.0
 
 # If a node is idle for this many minutes, it will be removed.
 idle_timeout_minutes: 5

python/ray/autoscaler/aws/example-full.yaml

@@ -9,15 +9,11 @@ min_workers: 0
 # node. This takes precedence over min_workers.
 max_workers: 2
 
-# The initial number of worker nodes to launch in addition to the head
-# node. When the cluster is first brought up (or when it is refreshed with a
-# subsequent `ray up`) this number of nodes will be started.
-initial_workers: 0
-
-# Whether or not to autoscale aggressively. If this is enabled, if at any point
-#   we would start more workers, we start at least enough to bring us to
-#   initial_workers.
-autoscaling_mode: default
+# The autoscaler will scale up the cluster faster with higher upscaling speed.
+# E.g., if the task requires adding more nodes then autoscaler will gradually
+# scale up the cluster in chunks of upscaling_speed*currently_running_nodes.
+# This number should be > 0.
+upscaling_speed: 1.0
 
 # This executes all commands on all nodes in the docker container,
 # and opens all the necessary ports to support the Ray cluster.
@@ -37,13 +33,6 @@ docker:
     # worker_image: "rayproject/ray:latest-cpu"
     # worker_run_options: []
 
-# The autoscaler will scale up the cluster to this target fraction of resource
-# usage. For example, if a cluster of 10 nodes is 100% busy and
-# target_utilization is 0.8, it would resize the cluster to 13. This fraction
-# can be decreased to increase the aggressiveness of upscaling.
-# This max value allowed is 1.0, which is the most conservative setting.
-target_utilization_fraction: 0.8
-
 # If a node is idle for this many minutes, it will be removed.
 idle_timeout_minutes: 5
 

python/ray/autoscaler/aws/example-gpu-docker.yaml

@@ -9,15 +9,11 @@ min_workers: 0
 # node. This takes precedence over min_workers.
 max_workers: 2
 
-# The initial number of worker nodes to launch in addition to the head
-# node. When the cluster is first brought up (or when it is refreshed with a
-# subsequent `ray up`) this number of nodes will be started.
-initial_workers: 0
-
-# Whether or not to autoscale aggressively. If this is enabled, if at any point
-#   we would start more workers, we start at least enough to bring us to
-#   initial_workers.
-autoscaling_mode: default
+# The autoscaler will scale up the cluster faster with higher upscaling speed.
+# E.g., if the task requires adding more nodes then autoscaler will gradually
+# scale up the cluster in chunks of upscaling_speed*currently_running_nodes.
+# This number should be > 0.
+upscaling_speed: 1.0
 
 # This executes all commands on all nodes in the docker container,
 # and opens all the necessary ports to support the Ray cluster.
@@ -31,13 +27,6 @@ docker:
 
     # worker_image: "rayproject/ray:latest"
 
-# The autoscaler will scale up the cluster to this target fraction of resource
-# usage. For example, if a cluster of 10 nodes is 100% busy and
-# target_utilization is 0.8, it would resize the cluster to 13. This fraction
-# can be decreased to increase the aggressiveness of upscaling.
-# This value must be less than 1.0 for scaling to happen.
-target_utilization_fraction: 0.8
-
 # If a node is idle for this many minutes, it will be removed.
 idle_timeout_minutes: 5
 

python/ray/autoscaler/aws/example-ml.yaml

@@ -14,15 +14,11 @@ min_workers: 0
 # node. This takes precedence over min_workers.
 max_workers: 0
 
-# The initial number of worker nodes to launch in addition to the head
-# node. When the cluster is first brought up (or when it is refreshed with a
-# subsequent `ray up`) this number of nodes will be started.
-initial_workers: 0
-
-# Whether or not to autoscale aggressively. If this is enabled, if at any point
-#   we would start more workers, we start at least enough to bring us to
-#   initial_workers.
-autoscaling_mode: default
+# The autoscaler will scale up the cluster faster with higher upscaling speed.
+# E.g., if the task requires adding more nodes then autoscaler will gradually
+# scale up the cluster in chunks of upscaling_speed*currently_running_nodes.
+# This number should be > 0.
+upscaling_speed: 1.0
 
 # This executes all commands on all nodes in the docker container,
 # and opens all the necessary ports to support the Ray cluster.
@@ -40,13 +36,6 @@ docker:
 
     # worker_image: "rayproject/ray:latest"
 
-# The autoscaler will scale up the cluster to this target fraction of resource
-# usage. For example, if a cluster of 10 nodes is 100% busy and
-# target_utilization is 0.8, it would resize the cluster to 13. This fraction
-# can be decreased to increase the aggressiveness of upscaling.
-# This max value allowed is 1.0, which is the most conservative setting.
-target_utilization_fraction: 0.8
-
 # If a node is idle for this many minutes, it will be removed.
 idle_timeout_minutes: 5
 

python/ray/autoscaler/aws/example-multi-node-type.yaml

@@ -3,6 +3,12 @@ cluster_name: multi_node_type
 min_workers: 1
 max_workers: 40
 
+# The autoscaler will scale up the cluster faster with higher upscaling speed.
+# E.g., if the task requires adding more nodes then autoscaler will gradually
+# scale up the cluster in chunks of upscaling_speed*currently_running_nodes.
+# This number should be > 0.
+upscaling_speed: 1.0
+
 # Cloud-provider specific configuration.
 provider:
     type: aws
@@ -62,8 +68,6 @@ head_node:
 worker_nodes:
     ImageId: latest_dlami
 
-# Configure the cluster for very conservative auto-scaling otherwise.
-target_utilization_fraction: 1.0
 idle_timeout_minutes: 2
 
 # How Ray will authenticate with newly launched nodes.