Skip to content

Commit cfd3224

Browse files
committed
Add 335 docs
1 parent ed643ba commit cfd3224

File tree

778 files changed

+109676
-2
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

778 files changed

+109676
-2
lines changed
48.1 KB
Loading

335/_sources/admin.rst.txt

Lines changed: 15 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,15 @@
1+
**************
2+
Administration
3+
**************
4+
5+
.. toctree::
6+
:maxdepth: 1
7+
8+
admin/web-interface
9+
admin/tuning
10+
admin/jmx
11+
admin/properties
12+
admin/spill
13+
admin/resource-groups
14+
admin/session-property-managers
15+
admin/dist-sort
Lines changed: 17 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,17 @@
1+
================
2+
Distributed Sort
3+
================
4+
5+
Distributed sort allows to sort data, which exceeds ``query.max-memory-per-node``.
6+
Distributed sort is enabled via the ``distributed_sort`` session property, or
7+
``distributed-sort`` configuration property set in
8+
``etc/config.properties`` of the coordinator. Distributed sort is enabled by
9+
default.
10+
11+
When distributed sort is enabled, the sort operator executes in parallel on multiple
12+
nodes in the cluster. Partially sorted data from each Presto worker node is then streamed
13+
to a single worker node for a final merge. This technique allows to utilize memory of multiple
14+
Presto worker nodes for sorting. The primary purpose of distributed sort is to allow for sorting
15+
of data sets which don't normally fit into single node memory. Performance improvement
16+
can be expected, but it won't scale linearly with the number of nodes, since the
17+
data needs to be merged by a single node.

335/_sources/admin/jmx.rst.txt

Lines changed: 76 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,76 @@
1+
===================
2+
Monitoring with JMX
3+
===================
4+
5+
Presto exposes a large number of different metrics via the Java Management Extensions (JMX).
6+
7+
You have to enable JMX by setting the ports used by the RMI registry and server
8+
in the :ref:`config.properties file <config_properties>`:
9+
10+
.. code-block:: none
11+
12+
jmx.rmiregistry.port=9080
13+
jmx.rmiserver.port=9081
14+
15+
* ``jmx.rmiregistry.port``:
16+
Specifies the port for the JMX RMI registry. JMX clients should connect to this port.
17+
18+
* ``jmx.rmiserver.port``:
19+
Specifies the port for the JMX RMI server. Presto exports many metrics,
20+
that are useful for monitoring via JMX.
21+
22+
JConsole (supplied with the JDK), `VisualVM <https://visualvm.github.io/>`_, and
23+
many other tools can be used to access the metrics in a client application.
24+
Many monitoring solutions support JMX. You can also use the
25+
:doc:`/connector/jmx` and query the metrics using SQL.
26+
27+
Many of these JMX metrics are a complex metric object such as a ``CounterStat``
28+
that has a collection of related metrics. For example, ``InputPositions`` has
29+
``InputPositions.TotalCount``, ``InputPositions.OneMinute.Count``, and so on.
30+
31+
A small subset of the available metrics are described below.
32+
33+
JVM
34+
---
35+
36+
* Heap size: ``java.lang:type=Memory:HeapMemoryUsage.used``
37+
* Thread count: ``java.lang:type=Threading:ThreadCount``
38+
39+
Presto Cluster and Nodes
40+
------------------------
41+
42+
* Active nodes:
43+
``presto.failureDetector:name=HeartbeatFailureDetector:ActiveCount``
44+
45+
* Free memory (general pool):
46+
``presto.memory:type=ClusterMemoryPool:name=general:FreeDistributedBytes``
47+
48+
* Cumulative count (since Presto started) of queries that ran out of memory and were killed:
49+
``presto.memory:name=ClusterMemoryManager:QueriesKilledDueToOutOfMemory``
50+
51+
Presto Queries
52+
--------------
53+
54+
* Active queries currently executing or queued: ``presto.execution:name=QueryManager:RunningQueries``
55+
56+
* Queries started: ``presto.execution:name=QueryManager:StartedQueries.FiveMinute.Count``
57+
58+
* Failed queries from last 5 min (all): ``presto.execution:name=QueryManager:FailedQueries.FiveMinute.Count``
59+
* Failed queries from last 5 min (internal): ``presto.execution:name=QueryManager:InternalFailures.FiveMinute.Count``
60+
* Failed queries from last 5 min (external): ``presto.execution:name=QueryManager:ExternalFailures.FiveMinute.Count``
61+
* Failed queries (user): ``presto.execution:name=QueryManager:UserErrorFailures.FiveMinute.Count``
62+
63+
* Execution latency (P50): ``presto.execution:name=QueryManager:ExecutionTime.FiveMinutes.P50``
64+
* Input data rate (P90): ``presto.execution:name=QueryManager:WallInputBytesRate.FiveMinutes.P90``
65+
66+
Presto Tasks
67+
------------
68+
69+
* Input data bytes: ``presto.execution:name=TaskManager:InputDataSize.FiveMinute.Count``
70+
* Input rows: ``presto.execution:name=TaskManager:InputPositions.FiveMinute.Count``
71+
72+
Connectors
73+
----------
74+
75+
Many connectors provide their own metrics. The metric names typically start with
76+
``presto.plugin``.

0 commit comments

Comments
 (0)