diff --git a/.github/workflows/README.md b/.github/workflows/README.md
index 87e124f97..713089ad9 100644
--- a/.github/workflows/README.md
+++ b/.github/workflows/README.md
@@ -14,7 +14,7 @@
 2. Click on Run workflow button
 3. Fill in the form and click on Run workflow button
    1. Name of the cluster - the name of the cluster that will be later used for other workflows. Default value is `gh-${{ github.repository_owner }}`, this results in `gh-<owner of fork>`.
-   2. Instance type for compute nodes - see [AWS EC2 instance types](https://aws.amazon.com/ec2/instance-types/). Default value is `m5.xlarge`.
+   2. Instance type for compute nodes - see [AWS EC2 instance types](https://aws.amazon.com/ec2/instance-types/). Default value is `m5.2xlarge`.
    3. Deploy to multiple availability zones in the region - if checked, the cluster will be deployed to multiple availability zones in the region. Default value is `false`.
    4. Number of worker nodes to provision - number of compute nodes in the cluster. Default value is `2`.
 4. Wait for the workflow to finish.
diff --git a/.github/workflows/keycloak-scalability-benchmark.yml b/.github/workflows/keycloak-scalability-benchmark.yml
index ff125210e..528ce44f5 100644
--- a/.github/workflows/keycloak-scalability-benchmark.yml
+++ b/.github/workflows/keycloak-scalability-benchmark.yml
@@ -93,7 +93,7 @@ jobs:
 
       - name: Allow cluster to scale
         if: ${{ !inputs.skipCreateDeployment }}
-        run: rosa edit machinepool -c ${{ inputs.clusterName }} --min-replicas 3 --autorepair scaling
+        run: rosa edit machinepool -c ${{ inputs.clusterName }} --min-replicas 4 --autorepair scaling
 
       - name: Create Keycloak deployment
         if: ${{ !inputs.skipCreateDeployment }}
diff --git a/.github/workflows/rosa-cluster-create.yml b/.github/workflows/rosa-cluster-create.yml
index 5b2452f07..41ffb09d8 100644
--- a/.github/workflows/rosa-cluster-create.yml
+++ b/.github/workflows/rosa-cluster-create.yml
@@ -11,6 +11,7 @@ on:
         type: string
       computeMachineType:
         description: 'Instance type for the compute nodes'
+        default: 'm5.2xlarge'
         type: string
       availabilityZones:
         description: 'Availability zones to deploy to'
@@ -34,6 +35,7 @@ on:
         default: 10.0.0.0/24
       computeMachineType:
         description: 'Instance type for the compute nodes'
+        default: 'm5.2xlarge'
         type: string
       availabilityZones:
         description: 'Availability zones to deploy to'
diff --git a/.github/workflows/rosa-multi-az-cluster-create.yml b/.github/workflows/rosa-multi-az-cluster-create.yml
index 0248b094d..a4ef689fe 100644
--- a/.github/workflows/rosa-multi-az-cluster-create.yml
+++ b/.github/workflows/rosa-multi-az-cluster-create.yml
@@ -141,8 +141,8 @@ jobs:
 
       - name: Scale ROSA clusters
         run: |
-          rosa edit machinepool -c ${{ env.CLUSTER_PREFIX }}-a --min-replicas 3 --max-replicas 10 --autorepair scaling
-          rosa edit machinepool -c ${{ env.CLUSTER_PREFIX }}-b --min-replicas 3 --max-replicas 10 --autorepair scaling
+          rosa edit machinepool -c ${{ env.CLUSTER_PREFIX }}-a --min-replicas 4 --max-replicas 15 --autorepair scaling
+          rosa edit machinepool -c ${{ env.CLUSTER_PREFIX }}-b --min-replicas 4 --max-replicas 15 --autorepair scaling
 
       - name: Setup Go Task
         uses: ./.github/actions/task-setup
diff --git a/doc/benchmark/modules/ROOT/pages/report/rosa-benchmark-key-results.adoc b/doc/benchmark/modules/ROOT/pages/report/rosa-benchmark-key-results.adoc
index b8563bbe7..2e4df1269 100644
--- a/doc/benchmark/modules/ROOT/pages/report/rosa-benchmark-key-results.adoc
+++ b/doc/benchmark/modules/ROOT/pages/report/rosa-benchmark-key-results.adoc
@@ -14,7 +14,7 @@ Collecting the CPU usage for refreshing a token is currently performed manually
 This setup is run https://github.com/keycloak/keycloak-benchmark/blob/main/.github/workflows/rosa-cluster-auto-provision-on-schedule.yml[daily on a GitHub action schedule]:
 
 * OpenShift 4.15.x deployed on AWS via ROSA with two AWS availability zones in AWS one region.
-* Machinepool with `m5.4xlarge` instances.
+* Machinepool with `m5.2xlarge` instances.
 * Keycloak 25 release candidate build deployed with Operator and 3 pods in each site as an active/passive setup, and Infinispan connecting the two sites.
 * Default user password hashing with Argon2 and 5 hash iterations and minimum memory size 7 MiB https://cheatsheetseries.owasp.org/cheatsheets/Password_Storage_Cheat_Sheet.html#argon2id[as recommended by OWASP].
 * Database seeded with 100,000 users and 100,000 clients.
@@ -53,7 +53,7 @@ Each test ran for 10 minutes.
 +
 [source,bash,subs="+quotes"]
 ----
-rosa edit machinepool -c  **<clustername>** --min-replicas 3 --autorepair scaling
+rosa edit machinepool -c  **<clustername>** --min-replicas 4 --autorepair scaling
 ----
 . Deploy Keycloak and Monitoring
 +
diff --git a/doc/kubernetes/modules/ROOT/pages/prerequisite/prerequisite-rosa.adoc b/doc/kubernetes/modules/ROOT/pages/prerequisite/prerequisite-rosa.adoc
index 465b23ffe..2ea41a975 100644
--- a/doc/kubernetes/modules/ROOT/pages/prerequisite/prerequisite-rosa.adoc
+++ b/doc/kubernetes/modules/ROOT/pages/prerequisite/prerequisite-rosa.adoc
@@ -36,7 +36,7 @@ After the installation process is finished, it creates a new admin user.
 CLUSTER_NAME=rosa-kcb
 VERSION=4.13.8
 REGION=eu-central-1
-COMPUTE_MACHINE_TYPE=m5.xlarge
+COMPUTE_MACHINE_TYPE=m5.2xlarge
 MULTI_AZ=false
 REPLICAS=3
 ----
@@ -85,14 +85,14 @@ The above installation script creates an admin user automatically but in case th
 == Scaling the cluster's nodes on demand
 
 The standard setup of nodes might be too small for running a load test, at the same time using a different instance type and rebuilding the cluster takes a lot of time (about 45 minutes).
-To scale the cluster on demand, the standard setup has a machine pool named `scaling` with instances of type `m5.4xlarge` which is auto-scaled based on the current demand from 0 to 10 instances.
+To scale the cluster on demand, the standard setup has a machine pool named `scaling` with instances of type `m5.2xlarge` which is auto-scaled based on the current demand from 4 to 15 instances.
 However, auto-scaling of worker nodes is quite time-consuming as nodes are scaled one by one.
 
 To scale the machine pool faster, issue a command like the following, and the additional nodes will be spawned at the same time:
 
 [source,bash,subs=+quotes]
 ----
-rosa edit machinepool -c _**clustername**_ --min-replicas 3 --autorepair scaling
+rosa edit machinepool -c _**clustername**_ --min-replicas 4 --autorepair scaling
 ----
 
 To allow scaling the machine pool back to 0, use a command like the following:
diff --git a/provision/aws/rosa_create_cluster.sh b/provision/aws/rosa_create_cluster.sh
index 97e91bf00..2b0e9ba72 100755
--- a/provision/aws/rosa_create_cluster.sh
+++ b/provision/aws/rosa_create_cluster.sh
@@ -63,7 +63,7 @@ fi
 
 SCALING_MACHINE_POOL=$(rosa list machinepools -c "${CLUSTER_NAME}" -o json | jq -r '.[] | select(.id == "scaling") | .id')
 if [[ "${SCALING_MACHINE_POOL}" != "scaling" ]]; then
-    rosa create machinepool -c "${CLUSTER_NAME}" --instance-type m5.4xlarge --max-replicas 10 --min-replicas 1 --name scaling --enable-autoscaling --autorepair
+    rosa create machinepool -c "${CLUSTER_NAME}" --instance-type "${COMPUTE_MACHINE_TYPE:-m5.2xlarge}" --max-replicas 15 --min-replicas 1 --name scaling --enable-autoscaling --autorepair
 fi
 
 cd ${SCRIPT_DIR}
diff --git a/provision/opentofu/modules/rosa/hcp/variables.tf b/provision/opentofu/modules/rosa/hcp/variables.tf
index bff117433..b21345085 100644
--- a/provision/opentofu/modules/rosa/hcp/variables.tf
+++ b/provision/opentofu/modules/rosa/hcp/variables.tf
@@ -61,7 +61,7 @@ variable "openshift_version" {
 
 variable "instance_type" {
   type     = string
-  default  = "m5.4xlarge"
+  default  = "m5.2xlarge"
   nullable = false
 }