Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Switch to using pd-balanced for all user & dask nodes #1124

Merged
merged 2 commits into from
Mar 18, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 13 additions & 4 deletions terraform/gcp/cluster.tf
Original file line number Diff line number Diff line change
Expand Up @@ -194,6 +194,13 @@ resource "google_container_node_pool" "notebook" {


node_config {

# Balanced disks are much faster than standard disks, and much cheaper
# than SSD disks. It contributes heavily to how fast new nodes spin up,
# as images being pulled takes up a lot of new node spin up time.
# Faster disks provide faster image pulls!
disk_type = "pd-balanced"

workload_metadata_config {
# Config Connector requires workload identity to be enabled (via GKE_METADATA_SERVER).
# If config connector is not necessary, we use simple metadata concealment
Expand Down Expand Up @@ -257,10 +264,12 @@ resource "google_container_node_pool" "dask_worker" {
node_config {

preemptible = true
# SSD Disks for dask workers make image pulls much faster
# Since we might have many dask workers spinning up at the
# same time, the extra cost of using this is probably worth it.
disk_type = "pd-ssd"

# Balanced disks are much faster than standard disks, and much cheaper
# than SSD disks. It contributes heavily to how fast new nodes spin up,
# as images being pulled takes up a lot of new node spin up time.
# Faster disks provide faster image pulls!
disk_type = "pd-balanced"

workload_metadata_config {
# Config Connector requires workload identity to be enabled (via GKE_METADATA_SERVER).
Expand Down
4 changes: 2 additions & 2 deletions terraform/gcp/projects/cloudbank.tfvars
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ notebook_nodes = {
min : 0,
max : 20,
machine_type : "n1-highmem-4",
labels: {}
labels: {},
},
}

Expand All @@ -24,7 +24,7 @@ dask_nodes = {
min : 0,
max : 100,
machine_type : "n1-highmem-4",
labels: {}
labels: {},
},
}

Expand Down
20 changes: 10 additions & 10 deletions terraform/gcp/projects/meom-ige.tfvars
Original file line number Diff line number Diff line change
Expand Up @@ -24,31 +24,31 @@ notebook_nodes = {
min : 0,
max : 20,
machine_type : "n1-standard-2",
labels: {}
labels: {},
},
"medium" : {
min : 0,
max : 20,
machine_type : "n1-standard-8",
labels: {}
labels: {},
},
"large" : {
min : 0,
max : 20,
machine_type : "n1-standard-16",
labels: {}
labels: {},
},
"very-large" : {
min : 0,
max : 20,
machine_type : "n1-standard-32",
labels: {}
labels: {},
},
"huge" : {
min : 0,
max : 20,
machine_type : "n1-standard-64",
labels: {}
labels: {},
},

}
Expand All @@ -58,31 +58,31 @@ dask_nodes = {
min : 0,
max : 20,
machine_type : "n1-standard-2",
labels: {}
labels: {},
},
"medium" : {
min : 0,
max : 20,
machine_type : "n1-standard-8",
labels: {}
labels: {},
},
"large" : {
min : 0,
max : 20,
machine_type : "n1-standard-16",
labels: {}
labels: {},
},
"very-large" : {
min : 0,
max : 20,
machine_type : "n1-standard-32",
labels: {}
labels: {},
},
"huge" : {
min : 0,
max : 20,
machine_type : "n1-standard-64",
labels: {}
labels: {},
},

}
Expand Down
12 changes: 6 additions & 6 deletions terraform/gcp/projects/pangeo-hubs.tfvars
Original file line number Diff line number Diff line change
Expand Up @@ -23,19 +23,19 @@ notebook_nodes = {
min : 0,
max : 100,
machine_type : "n1-standard-4",
labels: {}
labels: {},
},
"medium" : {
min : 0,
max : 100,
machine_type : "n1-standard-8",
labels: {}
labels: {},
},
"large" : {
min : 0,
max : 100,
machine_type : "n1-standard-16",
labels: {}
labels: {},
},
}

Expand All @@ -44,18 +44,18 @@ dask_nodes = {
min : 0,
max : 100,
machine_type : "n1-standard-4",
labels: {}
labels: {},
},
"medium" : {
min : 0,
max : 100,
machine_type : "n1-standard-8",
labels: {}
labels: {},
},
"large" : {
min : 0,
max : 100,
machine_type : "n1-standard-16",
labels: {}
labels: {},
},
}
4 changes: 2 additions & 2 deletions terraform/gcp/projects/pilot-hubs.tfvars
Original file line number Diff line number Diff line change
Expand Up @@ -14,8 +14,8 @@ notebook_nodes = {
min : 0,
max : 20,
machine_type : "n1-highmem-4",
labels: { }
}
labels: { },
},
}

dask_nodes = {
Expand Down
16 changes: 8 additions & 8 deletions terraform/gcp/variables.tf
Original file line number Diff line number Diff line change
Expand Up @@ -189,8 +189,8 @@ variable "enable_private_cluster" {
}

variable "enable_filestore" {
type = bool
default = false
type = bool
default = false
description = <<-EOT
Deploy a Google FileStore for home directories
Expand All @@ -201,8 +201,8 @@ variable "enable_filestore" {
}

variable "filestore_capacity_gb" {
type = number
default = 1024
type = number
default = 1024
description = <<-EOT
Minimum size (in GB) of Google FileStore.
Expand All @@ -211,8 +211,8 @@ variable "filestore_capacity_gb" {
}

variable "filestore_tier" {
type = string
default = "BASIC_HDD"
type = string
default = "BASIC_HDD"
description = <<-EOT
Google FileStore service tier to use.
Expand All @@ -222,8 +222,8 @@ variable "filestore_tier" {
}

variable "enable_node_autoprovisioning" {
type = bool
default = false
type = bool
default = false
description = <<-EOT
Enable auto-provisioning of nodes based on workload
EOT
Expand Down