benchmarking: Add role to create benchmark hosts

Basic benchmarking will require two hosts, a Gitaly instance and a
client instance to send traffic from. Gitaly Cluster is beyond the scope
of this initial effort.

Create a terraform job that creates both hosts, with port 8075 open on
the Gitaly node for traffic. We use a `t2d` instance for Gitaly as these
provide 4 physical cores, as opposed to 2 hyperthreaded cores. In theory
this could reduce performance jitter, though I have not measured this to
be sure.

A disk image containing the test repositories is attached to the Gitaly
node on creation. These repositories are:

- git.git - A smaller repository with a fair amount of history.
- gitlab.git - Uses an object pool and has ~6,000,000 refs.
- linux.git - A large and well-groomed repository.
- homebrew-core.git - Has very large trees.
- chromium.git - Extremely large (40 GiB), with ~4,000,000 refs.

This task borrows its structure from the old Gitaly Cluster demo script in
`_support/terraform`.

9 files changed, 202 insertions, 0 deletions

diff --git a/_support/benchmarking/.gitignore b/_support/benchmarking/.gitignore
new file mode 100644
index 000000000..aa8762020
--- /dev/null
+++ b/_support/benchmarking/.gitignore
@@ -0,0 +1,4 @@
+/hosts.ini
+/terraform/*
+!/terraform/main.tf
+/group_vars/all.yml
diff --git a/_support/benchmarking/create-benchmark-instance b/_support/benchmarking/create-benchmark-instance
new file mode 100755
index 000000000..0173ac095
--- /dev/null
+++ b/_support/benchmarking/create-benchmark-instance
@@ -0,0 +1,2 @@
+#!/bin/sh
+exec ansible-playbook -l localhost create.yml "$@"
diff --git a/_support/benchmarking/create.yml b/_support/benchmarking/create.yml
new file mode 100644
index 000000000..1e4942ba2
--- /dev/null
+++ b/_support/benchmarking/create.yml
@@ -0,0 +1,21 @@
+---
+- name: Set user-defined variables
+  hosts: localhost
+  vars_prompt:
+    - name: gitaly_revision
+      prompt: "Enter Gitaly revision to build"
+      default: HEAD
+      private: false
+
+    - name: gitaly_benchmarking_instance_name
+      prompt: "Enter a name for your benchmark instance"
+      default: "{{ lookup('env', 'USER') }}-{{ lookup('pipe', 'date +%Y-%m-%d') }}"
+      private: false
+
+    - name: ssh_pubkey
+      prompt: "Enter the path to your SSH public key"
+      default: "{{ lookup('env', 'HOME') }}/.ssh/id_ed25519.pub"
+      private: false
+  roles:
+    - deploy
+  gather_facts: false
diff --git a/_support/benchmarking/roles/deploy/handlers/main.yml b/_support/benchmarking/roles/deploy/handlers/main.yml
new file mode 100644
index 000000000..d5d2716a9
--- /dev/null
+++ b/_support/benchmarking/roles/deploy/handlers/main.yml
@@ -0,0 +1,16 @@
+---
+- name: scan SSH keys
+  command: "ssh-keyscan {{ item }}"
+  register: ssh_keys
+  loop: "{{ [tfstate.outputs.gitaly_ssh_ip.value] | list + [tfstate.outputs.client_ssh_ip.value] }}"
+  listen: add hostkeys
+  retries: 15
+  until: ssh_keys is not failed
+  delay: 10
+
+- name: add SSH keys to known hosts
+  known_hosts:
+    name: "{{ item.item }}"
+    key: "{{ item.stdout }}"
+  loop: "{{ ssh_keys.results }}"
+  listen: add hostkeys
diff --git a/_support/benchmarking/roles/deploy/tasks/main.yml b/_support/benchmarking/roles/deploy/tasks/main.yml
new file mode 100644
index 000000000..fc167cd23
--- /dev/null
+++ b/_support/benchmarking/roles/deploy/tasks/main.yml
@@ -0,0 +1,38 @@
+---
+- name: GCloud login
+  block:
+    - name: GCloud login status
+      command: gcloud auth application-default print-access-token
+      changed_when: false
+  rescue:
+    - name: GCloud login
+      command: gcloud auth application-default login
+
+- name: terraform.tfvars
+  template:
+    src: terraform.tfvars.j2
+    dest: "{{ playbook_dir }}/terraform/terraform.tfvars"
+
+- name: terraform apply
+  terraform:
+    project_path: "{{ playbook_dir }}/terraform"
+    variables_file: terraform.tfvars
+    force_init: true
+  register: tfstate
+  notify:
+    - add hostkeys
+
+- name: hosts.ini
+  template:
+    src: hosts.ini.j2
+    dest: "{{ playbook_dir }}/hosts.ini"
+
+- name: Create group_vars directory
+  file:
+    path: "{{ playbook_dir }}/group_vars"
+    state: directory
+
+- name: all.yml
+  template:
+    src: all.yml.j2
+    dest: "{{ playbook_dir }}/group_vars/all.yml"
diff --git a/_support/benchmarking/roles/deploy/templates/all.yml.j2 b/_support/benchmarking/roles/deploy/templates/all.yml.j2
new file mode 100644
index 000000000..e6ae9a89b
--- /dev/null
+++ b/_support/benchmarking/roles/deploy/templates/all.yml.j2
@@ -0,0 +1,2 @@
+---
+gitaly_revision: "{{ gitaly_revision }}"
diff --git a/_support/benchmarking/roles/deploy/templates/hosts.ini.j2 b/_support/benchmarking/roles/deploy/templates/hosts.ini.j2
new file mode 100644
index 000000000..fb48fefa9
--- /dev/null
+++ b/_support/benchmarking/roles/deploy/templates/hosts.ini.j2
@@ -0,0 +1,5 @@
+[gitaly]
+{{ tfstate.outputs.gitaly_ssh_ip.value }} internal={{ tfstate.outputs.gitaly_internal_ip.value }} ansible_user=gitaly_bench
+
+[client]
+{{ tfstate.outputs.client_ssh_ip.value }} internal={{ tfstate.outputs.client_internal_ip.value }} ansible_user=gitaly_bench
diff --git a/_support/benchmarking/roles/deploy/templates/terraform.tfvars.j2 b/_support/benchmarking/roles/deploy/templates/terraform.tfvars.j2
new file mode 100644
index 000000000..e8c156c55
--- /dev/null
+++ b/_support/benchmarking/roles/deploy/templates/terraform.tfvars.j2
@@ -0,0 +1,3 @@
+# This variable will be prefixed to all machines created by terraform
+gitaly_benchmarking_instance_name = "{{ gitaly_benchmarking_instance_name }}"
+ssh_pubkey = "{{ lookup('file', ssh_pubkey) }}"
diff --git a/_support/benchmarking/terraform/main.tf b/_support/benchmarking/terraform/main.tf
new file mode 100644
index 000000000..59f701d95
--- /dev/null
+++ b/_support/benchmarking/terraform/main.tf
@@ -0,0 +1,111 @@
+variable "project" { default = "gitaly-benchmark-0150d6cf" }
+variable "benchmark_region" { default = "us-central1" }
+variable "benchmark_zone" { default = "us-central1-a" }
+variable "gitaly_benchmarking_instance_name" { }
+variable "ssh_pubkey" { }
+variable "os_image" { default = "ubuntu-os-cloud/ubuntu-2204-lts" }
+variable "startup_script" {
+  default = <<EOF
+    set -e
+    if [ -d /src/gitaly ] ; then exit; fi
+  EOF
+}
+variable "gitaly_machine_type" { default = "t2d-standard-4" }
+variable "client_machine_type" { default = "n1-standard-1" }
+variable "boot_disk_size" { default = "20" }
+
+provider "google" {
+  project = var.project
+  region  = var.benchmark_region
+  zone    = var.benchmark_zone
+}
+
+resource "google_compute_network" "default" {
+  name = "test-network"
+}
+
+data "google_compute_disk" "repository-disk" {
+  name = "git-repos"
+  project = "gitaly-benchmark-0150d6cf"
+}
+
+resource "google_compute_disk" "repository-disk" {
+  name = format("%s-repository-disk", var.gitaly_benchmarking_instance_name)
+  type = "pd-balanced"
+  image = format("projects/%s/global/images/git-repositories", var.project)
+}
+
+resource "google_compute_instance" "gitaly" {
+  name         = format("%s-gitaly", var.gitaly_benchmarking_instance_name)
+  machine_type = var.gitaly_machine_type
+
+  boot_disk {
+    initialize_params {
+      image = var.os_image
+      size = var.boot_disk_size
+    }
+  }
+
+  attached_disk {
+    source = google_compute_disk.repository-disk.name
+    device_name = "repository-disk"
+  }
+
+  network_interface {
+    network = "default"
+    subnetwork = "default"
+    access_config {}
+  }
+
+  metadata = {
+    ssh-keys = format("gitaly_bench:%s", var.ssh_pubkey)
+    startup-script = <<EOF
+      ${var.startup_script}
+    EOF
+  }
+
+  tags = ["gitaly"]
+
+  lifecycle {
+    ignore_changes = [attached_disk]
+  }
+}
+
+resource "google_compute_instance" "client" {
+  name         = format("%s-client", var.gitaly_benchmarking_instance_name)
+  machine_type = var.client_machine_type
+
+  boot_disk {
+    initialize_params {
+      image = var.os_image
+      size = var.boot_disk_size
+    }
+  }
+
+  network_interface {
+    subnetwork = "default"
+    access_config {}
+  }
+
+  metadata = {
+    ssh-keys = format("gitaly_bench:%s", var.ssh_pubkey)
+    startup-script = <<EOF
+      ${var.startup_script}
+    EOF
+  }
+}
+
+output "gitaly_internal_ip" {
+  value = google_compute_instance.gitaly.network_interface[0].network_ip
+}
+output "gitaly_ssh_ip" {
+  value = google_compute_instance.gitaly.network_interface[0].access_config[0].nat_ip
+}
+
+output "client_internal_ip" {
+  value = google_compute_instance.client.network_interface[0].network_ip
+}
+
+output "client_ssh_ip" {
+  value = google_compute_instance.client.network_interface[0].access_config[0].nat_ip
+}