Skip to content

Commit

Permalink
Clarify source of github_events table (#850)
Browse files Browse the repository at this point in the history
  • Loading branch information
begriffs authored and jonels-msft committed Aug 30, 2019
1 parent 62ed1a4 commit 2b9f357
Showing 1 changed file with 2 additions and 3 deletions.
5 changes: 2 additions & 3 deletions develop/reference_workarounds.rst
Original file line number Diff line number Diff line change
Expand Up @@ -72,7 +72,7 @@ Temp Tables: the Workaround of Last Resort

There are still a few queries that are :ref:`unsupported <unsupported>` even with the use of push-pull execution via subqueries. One of them is running window functions that partition by a non-distribution column.

Suppose we have a table called :code:`github_events`, distributed by the column :code:`user_id`. Then the following window function will not work:
In our :ref:`real-time analytics tutorial <real_time_analytics_tutorial>` we created a table called :code:`github_events`, distributed by the column :code:`user_id`. If we want to query it and count the events for a preselected of repos, one way to do it in SQL is with a window function. However, partitioning on a non-distribution column does not work:

.. code-block:: sql
Expand All @@ -83,7 +83,7 @@ Suppose we have a table called :code:`github_events`, distributed by the column
FROM github_events
WHERE repo_id IN (8514, 15435, 19438, 21692);
There is another trick though. We can pull the relevant information to the coordinator as a temporary table:
There is a trick, though. We can pull the relevant information to the coordinator as a temporary table:

.. code-block:: sql
Expand All @@ -102,4 +102,3 @@ There is another trick though. We can pull the relevant information to the coord
FROM results;
Creating a temporary table on the coordinator is a last resort. It is limited by the disk size and CPU of the node.

0 comments on commit 2b9f357

Please sign in to comment.