diff --git a/tests/storage/concurrent_vm_boot/__init__.py b/tests/storage/concurrent_vm_boot/__init__.py
new file mode 100644
index 0000000000..e69de29bb2
diff --git a/tests/storage/concurrent_vm_boot/test_concurrent_vm_boot.py b/tests/storage/concurrent_vm_boot/test_concurrent_vm_boot.py
new file mode 100644
index 0000000000..ef7a6a8ccc
--- /dev/null
+++ b/tests/storage/concurrent_vm_boot/test_concurrent_vm_boot.py
@@ -0,0 +1,58 @@
+"""
+Concurrent VM Boot Tests
+
+Validates booting 20 virtual machines simultaneously, each with a multi-disk
+configuration: one cloned boot volume, one cloud-init disk, and three blank
+data volumes (five disk devices total in the VMI spec).
+
+Jira: https://redhat.atlassian.net/browse/CNV-88906  # <skip-jira-utils-check>
+
+Markers:
+    - tier3
+    - conformance
+    - special_infra
+    - high_resource_vm
+"""
+
+import pytest
+
+__test__ = False
+
+pytestmark = [pytest.mark.tier3, pytest.mark.conformance, pytest.mark.special_infra, pytest.mark.high_resource_vm]
+
+
+class TestConcurrentVMBoot:
+    """
+    Tests for booting multiple VMs simultaneously with multi-disk configurations.
+
+    Preconditions:
+        - Fedora golden image DataSource available in the openshift-virtualization-os-images namespace
+        - All schedulable worker nodes share the same architecture as the golden image
+        - Storage class supporting dynamic provisioning and CSI volume cloning
+        - Sufficient cluster resources to schedule 20 VMs simultaneously
+    """
+
+    @pytest.mark.polarion("CNV-16335")
+    def test_concurrent_vms_boot_with_five_disks(self):
+        """
+        Test that 20 VMs boot simultaneously with five disk devices each and all reach Running state.
+
+        Preconditions:
+            - Fedora golden image DataSource available in the openshift-virtualization-os-images namespace
+            - All schedulable worker nodes share the same architecture as the golden image
+            - Storage class supporting dynamic provisioning and CSI volume cloning
+            - Sufficient cluster resources to schedule 20 VMs simultaneously
+            - 20 VMs created, each with one golden image boot volume (PVC clone via DataSource),
+              one cloud-init disk, and three blank data volumes
+            - All 20 VMs started simultaneously and reached Running state
+
+        Steps:
+            1. Create 20 VMs, each with one golden image boot volume (PVC clone via DataSource),
+               one cloud-init disk, and three blank data volumes
+            2. Start all 20 VMs simultaneously
+            3. Wait for all 20 VMs to reach Running state
+            4. For each VM, inspect the disk devices reported in the VMI spec
+
+        Expected:
+            - All 20 VMs report exactly five disk devices each in the VMI spec
+        """