diff --git a/CHANGELOG.md b/CHANGELOG.md
index 3b81aaf8ae..e8770573af 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -99,6 +99,10 @@
 - Add liveness probe to `keepalived` pod 
   (PR[#4118](https://github.com/scality/metalk8s/pull/4118))
 
+- Add `kubeReserved` and `systemReserved` resources allocation to `KubeletConfiguration`
+  following [Google's recommandations](https://cloud.google.com/kubernetes-engine/docs/concepts/plan-node-sizes#memory_and_cpu_reservations)
+  (PR[#4134](https://github.com/scality/metalk8s/pull/4134))
+
 ## Release 125.0.6 (In development)
 
 ### Enhancements
diff --git a/buildchain/buildchain/salt_tree.py b/buildchain/buildchain/salt_tree.py
index 93640042c7..f32338d313 100644
--- a/buildchain/buildchain/salt_tree.py
+++ b/buildchain/buildchain/salt_tree.py
@@ -667,6 +667,7 @@ def task(self) -> types.TaskDict:
     Path("salt/_modules/metalk8s_kubernetes_utils.py"),
     Path("salt/_modules/metalk8s_monitoring.py"),
     Path("salt/_modules/metalk8s_network.py"),
+    Path("salt/_modules/metalk8s_os.py"),
     Path("salt/_modules/metalk8s_package_manager_yum.py"),
     Path("salt/_modules/metalk8s_service_configuration.py"),
     Path("salt/_modules/metalk8s_solutions.py"),
diff --git a/salt/_modules/metalk8s_os.py b/salt/_modules/metalk8s_os.py
new file mode 100644
index 0000000000..470e5d469a
--- /dev/null
+++ b/salt/_modules/metalk8s_os.py
@@ -0,0 +1,59 @@
+"""
+MetalK8s OS module
+"""
+
+__virtualname__ = "metalk8s_os"
+
+
+def __virtual__():
+    return __virtualname__
+
+
+def get_kubereserved():
+    """Get kubeReserved memory and cpu allocations following Google's GKE recommandations.
+
+    For CPU resources, GKE reserves the following:
+    - 6% of the first core
+    - 1% of the next core (up to 2 cores)
+    - 0.5% of the next 2 cores (up to 4 cores)
+    - 0.25% of any cores above 4 cores
+
+    For memory resources, GKE reserves the following:
+    - 255 MiB of memory for machines with less than 1 GB of memory
+    - 25% of the first 4GB of memory
+    - 20% of the next 4GB of memory (up to 8GB)
+    - 10% of the next 8GB of memory (up to 16GB)
+    - 6% of the next 112GB of memory (up to 128GB)
+    - 2% of any memory above 128GB
+
+
+    https://cloud.google.com/kubernetes-engine/docs/concepts/plan-node-sizes#memory_and_cpu_reservations
+    https://learnk8s.io/allocatable-resources
+    """
+
+    os_cpu = __grains__["num_cpus"]
+
+    core_1 = 0.06
+    cores_2 = 0.01 * max([(min([os_cpu, 2]) - 1), 0])
+    cores_4 = 0.005 * max([(min([os_cpu, 4]) - 2), 0])
+    cores_above_4 = 0.0025 * max([(os_cpu - 4), 0])
+
+    kube_cpu = round((core_1 + cores_2 + cores_4 + cores_above_4) * 1000)
+
+    os_memory = __grains__["mem_total"]
+
+    gb_4 = 4 * 1024
+    gb_8 = 8 * 1024
+    gb_16 = 16 * 1024
+    gb_128 = 128 * 1024
+
+    memory_4 = 0.25 * min([os_memory, gb_4])
+    memory_8 = 0.2 * max([(min([os_memory, gb_8]) - gb_4), 0])
+    memory_16 = 0.1 * max([(min([os_memory, gb_16]) - gb_8), 0])
+    memory_128 = 0.06 * max([(min([os_memory, gb_128]) - gb_16), 0])
+    memory_above_128 = 0.02 * max([(os_memory - gb_128), 0])
+
+    kube_memory = round(memory_4 + memory_8 + memory_16 + memory_128 + memory_above_128)
+
+    # CPU is in millicores, memory is in MiB
+    return {"cpu": f"{kube_cpu}m", "memory": f"{kube_memory}Mi"}
diff --git a/salt/metalk8s/kubernetes/kubelet/standalone.sls b/salt/metalk8s/kubernetes/kubelet/standalone.sls
index 78ea7469c7..ec9ed92fe6 100644
--- a/salt/metalk8s/kubernetes/kubelet/standalone.sls
+++ b/salt/metalk8s/kubernetes/kubelet/standalone.sls
@@ -103,6 +103,10 @@ Create kubelet config file:
         {%- if pillar.get("kubernetes:kubelet:config:maxPods") %}
         maxPods: {{ pillar.kubernetes.kubelet.config.maxPods }}
         {%- endif %}
+        systemReserved:
+          cpu: 200m
+          memory: 200Mi
+        kubeReserved: {{ salt.metalk8s_os.get_kubereserved() | tojson }}
 {%- for key, value in kubelet.config.items() %}
         {{ key }}: {{ value }}
 {%- endfor %}
diff --git a/salt/tests/unit/formulas/fixtures/salt.py b/salt/tests/unit/formulas/fixtures/salt.py
index 705f2ab6c2..d95c004281 100644
--- a/salt/tests/unit/formulas/fixtures/salt.py
+++ b/salt/tests/unit/formulas/fixtures/salt.py
@@ -26,6 +26,7 @@
     "file_client": "remote",
 }
 
+
 # The "public methods" are dynamically added by the `register` decorator
 # pylint: disable=too-few-public-methods
 class SaltMock:
@@ -187,6 +188,7 @@ def register_basic(func_name: str) -> Callable[[MockFunc], MockFunc]:
 # }}}
 # Mock definitions {{{
 
+
 # Data-driven mocks {{{
 @register("config.get")
 def config_get(salt_mock: SaltMock, *args: Any, **kwargs: Any) -> Any:
@@ -548,5 +550,9 @@ def random_get_str(length: int = 20) -> str:
     return "".join(random.choices(allowed_chars, k=length))
 
 
+register_basic("metalk8s_os.get_kubereserved")(
+    MagicMock(return_value={"cpu": "100m", "memory": "1000Mi"})
+)
+
 # }}}
 # }}}
diff --git a/salt/tests/unit/modules/files/test_metalk8s_os.yaml b/salt/tests/unit/modules/files/test_metalk8s_os.yaml
new file mode 100644
index 0000000000..1f2b4a057b
--- /dev/null
+++ b/salt/tests/unit/modules/files/test_metalk8s_os.yaml
@@ -0,0 +1,67 @@
+get_kubereserved:
+  # 00 - 1 core and 2 GB of RAM
+  - num_cpus: 1
+    mem_total: 2048
+    result: 
+      cpu: 60m
+      memory: 512Mi
+  # 01 - 2 cores and 4 GB of RAM
+  - num_cpus: 2
+    mem_total: 4096
+    result: 
+      cpu: 70m
+      memory: 1024Mi
+  # 02 - 4 cores and 8 GB of RAM
+  - num_cpus: 4
+    mem_total: 8192
+    result: 
+      cpu: 80m
+      memory: 1843Mi
+  # 03 - 8 cores and 16 GB of RAM
+  - num_cpus: 8
+    mem_total: 16384
+    result: 
+      cpu: 90m
+      memory: 2662Mi
+  # 04 - 16 cores and 32 GB of RAM
+  - num_cpus: 16
+    mem_total: 32768
+    result: 
+      cpu: 110m
+      memory: 3645Mi
+  # 05 - 32 cores and 64 GB of RAM
+  - num_cpus: 32
+    mem_total: 65536
+    result: 
+      cpu: 150m
+      memory: 5612Mi
+  # 06 - 64 cores and 128 GB of RAM
+  - num_cpus: 64
+    mem_total: 131072
+    result: 
+      cpu: 230m
+      memory: 9544Mi
+  # 07 - 128 cores and 256 GB of RAM
+  - num_cpus: 128
+    mem_total: 262144
+    result: 
+      cpu: 390m
+      memory: 12165Mi
+  # 08 - 256 cores and 512 GB of RAM
+  - num_cpus: 256
+    mem_total: 524288
+    result: 
+      cpu: 710m
+      memory: 17408Mi
+  # 09 - 512 cores and 1024 GB of RAM
+  - num_cpus: 512
+    mem_total: 1048576
+    result: 
+      cpu: 1350m
+      memory: 27894Mi
+  # 10 - 3 cores and 14.2 GB of RAM
+  - num_cpus: 3
+    mem_total: 14540
+    result: 
+      cpu: 75m
+      memory: 2478Mi
\ No newline at end of file
diff --git a/salt/tests/unit/modules/test_metalk8s_os.py b/salt/tests/unit/modules/test_metalk8s_os.py
new file mode 100644
index 0000000000..ba2b6890c9
--- /dev/null
+++ b/salt/tests/unit/modules/test_metalk8s_os.py
@@ -0,0 +1,45 @@
+import os.path
+import yaml
+
+from unittest import TestCase
+from unittest.mock import MagicMock, patch
+
+from _modules import metalk8s_os
+
+from tests.unit import mixins
+from tests.unit import utils
+
+
+YAML_TESTS_FILE = os.path.join(
+    os.path.dirname(os.path.abspath(__file__)), "files", "test_metalk8s_os.yaml"
+)
+with open(YAML_TESTS_FILE) as fd:
+    YAML_TESTS_CASES = yaml.safe_load(fd)
+
+
+class Metalk8sOsTestCase(TestCase, mixins.LoaderModuleMockMixin):
+    """
+    TestCase for `metalk8s_os` module
+    """
+
+    loader_module = metalk8s_os
+
+    def test_virtual(self):
+        """
+        Tests the return of `__virtual__` function
+        """
+        self.assertEqual(metalk8s_os.__virtual__(), "metalk8s_os")
+
+    @utils.parameterized_from_cases(YAML_TESTS_CASES["get_kubereserved"])
+    def test_get_kubereserved(self, num_cpus, mem_total, result):
+        """
+        Tests the return of `get_kubereserved` function
+        """
+
+        grains_dict = {"num_cpus": num_cpus, "mem_total": mem_total}
+
+        with patch.dict(metalk8s_os.__grains__, grains_dict):
+            self.assertEqual(
+                metalk8s_os.get_kubereserved(),
+                result,
+            )