diff --git a/doc/source/tune-config.rst b/doc/source/tune-config.rst
index f00538421..71e23e08d 100644
--- a/doc/source/tune-config.rst
+++ b/doc/source/tune-config.rst
@@ -69,6 +69,30 @@ By default, each random variable and grid search point is sampled once. To take
 
 For more information on variant generation, see `variant_generator.py <https://github.com/ray-project/ray/blob/master/python/ray/tune/variant_generator.py>`__.
 
+Handling Large Datasets
+-----------------------
+
+You often will want to compute a large object (e.g., training data, model weights) on the driver and use that object within each trial. Ray Tune provides a ``pin_in_object_store`` utility function that can be used to broadcast such large objects. Objects pinned in this way will never be evicted from the Ray object store while the driver process is running, and can be efficiently retrieved from any task via ``get_pinned_object``.
+
+.. code-block:: python
+
+    import ray
+    from ray.tune import register_trainable, run_experiments
+    from ray.tune.util import pin_in_object_store, get_pinned_object
+
+    import numpy as np
+
+    ray.init()
+
+    # X_id can be referenced in closures
+    X_id = pin_in_object_store(np.random.random(size=100000000))
+
+    def f(config, reporter):
+        X = get_pinned_object(X_id)
+        # use X
+
+    register_trainable("f", f)
+    run_experiments(...)
 
 Resource Allocation
 -------------------
diff --git a/python/ray/tune/test/trial_runner_test.py b/python/ray/tune/test/trial_runner_test.py
index a45be3104..da0cfd5c0 100644
--- a/python/ray/tune/test/trial_runner_test.py
+++ b/python/ray/tune/test/trial_runner_test.py
@@ -13,6 +13,7 @@ from ray.tune import Trainable, TuneError
 from ray.tune import register_env, register_trainable, run_experiments
 from ray.tune.registry import _default_registry, TRAINABLE_CLASS
 from ray.tune.result import DEFAULT_RESULTS_DIR
+from ray.tune.util import pin_in_object_store, get_pinned_object
 from ray.tune.experiment import Experiment
 from ray.tune.trial import Trial, Resources
 from ray.tune.trial_runner import TrialRunner
@@ -28,6 +29,15 @@ class TrainableFunctionApiTest(unittest.TestCase):
         ray.worker.cleanup()
         _register_all()  # re-register the evicted objects
 
+    def testPinObject(self):
+        X = pin_in_object_store("hello")
+
+        @ray.remote
+        def f():
+            return get_pinned_object(X)
+
+        self.assertEqual(ray.get(f.remote()), "hello")
+
     def testRegisterEnv(self):
         register_env("foo", lambda: None)
         self.assertRaises(TypeError, lambda: register_env("foo", 2))
diff --git a/python/ray/tune/util.py b/python/ray/tune/util.py
new file mode 100644
index 000000000..c3d58cb47
--- /dev/null
+++ b/python/ray/tune/util.py
@@ -0,0 +1,33 @@
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import base64
+
+import ray
+from ray.tune.registry import _to_pinnable, _from_pinnable
+
+
+_pinned_objects = []
+PINNED_OBJECT_PREFIX = "ray.tune.PinnedObject:"
+
+
+def pin_in_object_store(obj):
+    obj_id = ray.put(_to_pinnable(obj))
+    _pinned_objects.append(ray.get(obj_id))
+    return "{}{}".format(
+        PINNED_OBJECT_PREFIX, base64.b64encode(obj_id.id()).decode("utf-8"))
+
+
+def get_pinned_object(pinned_id):
+    from ray.local_scheduler import ObjectID
+    return _from_pinnable(ray.get(ObjectID(
+        base64.b64decode(pinned_id[len(PINNED_OBJECT_PREFIX):]))))
+
+
+if __name__ == '__main__':
+    ray.init()
+    X = pin_in_object_store("hello")
+    print(X)
+    result = get_pinned_object(X)
+    print(result)