From 07367533de68817e1e6cf9cf2b056a04dd160c8a Mon Sep 17 00:00:00 2001
From: Eric Liang <ekl@databricks.com>
Date: Mon, 20 Jun 2016 21:56:44 -0700
Subject: [SPARK-16025][CORE] Document OFF_HEAP storage level in 2.0

This has changed from 1.6, and now stores memory off-heap using spark's off-heap support instead of in tachyon.

Author: Eric Liang <ekl@databricks.com>

Closes #13744 from ericl/spark-16025.
---
 docs/programming-guide.md | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/docs/programming-guide.md b/docs/programming-guide.md
index 97bcb511b2..3872aecff2 100644
--- a/docs/programming-guide.md
+++ b/docs/programming-guide.md
@@ -1220,6 +1220,11 @@ storage levels is:
   <td> MEMORY_ONLY_2, MEMORY_AND_DISK_2, etc.  </td>
   <td> Same as the levels above, but replicate each partition on two cluster nodes. </td>
 </tr>
+<tr>
+  <td> OFF_HEAP (experimental) </td>
+  <td> Similar to MEMORY_ONLY_SER, but store the data in
+    <a href="configuration.html#memory-management">off-heap memory</a>. This requires off-heap memory to be enabled. </td>
+</tr>
 </table>
 
 **Note:** *In Python, stored objects will always be serialized with the [Pickle](https://docs.python.org/2/library/pickle.html) library, 
-- 
cgit v1.2.3