diff --git a/changelog/5516.trivial.rst b/changelog/5516.trivial.rst
new file mode 100644
index 000000000..2f6b4e35e
--- /dev/null
+++ b/changelog/5516.trivial.rst
@@ -0,0 +1 @@
+Cache node splitting function which can improve collection performance in very large test suites.
diff --git a/src/_pytest/nodes.py b/src/_pytest/nodes.py
index 7e1c40bcb..9b78dca38 100644
--- a/src/_pytest/nodes.py
+++ b/src/_pytest/nodes.py
@@ -1,5 +1,6 @@
 import os
 import warnings
+from functools import lru_cache
 
 import py
 
@@ -13,6 +14,7 @@ SEP = "/"
 tracebackcutdir = py.path.local(_pytest.__file__).dirpath()
 
 
+@lru_cache(maxsize=None)
 def _splitnode(nodeid):
     """Split a nodeid into constituent 'parts'.
 
@@ -30,11 +32,12 @@ def _splitnode(nodeid):
     """
     if nodeid == "":
         # If there is no root node at all, return an empty list so the caller's logic can remain sane
-        return []
+        return ()
     parts = nodeid.split(SEP)
     # Replace single last element 'test_foo.py::Bar' with multiple elements 'test_foo.py', 'Bar'
     parts[-1:] = parts[-1].split("::")
-    return parts
+    # Convert parts into a tuple to avoid possible errors with caching of a mutable type
+    return tuple(parts)
 
 
 def ischildnode(baseid, nodeid):