提交 a3191d64 编写于 作者: L Lisa Owen 提交者: David Yozie

docs - discuss pxf fragment metadata caching (#7548)

* docs - discuss pxf fragment metadata caching

* a large number of
上级 1ac3dd8e
......@@ -140,7 +140,7 @@ Perform the following procedure to increase the heap size for the PXF agent runn
$ ssh gpadmin@<gpmaster>
```
2. Recall the location of the PXF user configuration directory ($PXF_CONF). Edit the `$PXF_CONF/conf/pxf-env.sh` file. For example:
2. Recall the location of the PXF user configuration directory (`$PXF_CONF`). Edit the `$PXF_CONF/conf/pxf-env.sh` file. For example:
``` shell
gpadmin@gpmaster$ vi $PXF_CONF/conf/pxf-env.sh
......@@ -152,6 +152,9 @@ Perform the following procedure to increase the heap size for the PXF agent runn
PXF_JVM_OPTS="-Xmx3g -Xms3g"
```
3. Save the file and exit the editor.
4. Use the `pxf cluster sync` command to copy the updated `pxf-env.sh` file to the Greenplum Database cluster. For example:
``` shell
......@@ -219,3 +222,37 @@ You can use the `PXF_JVM_OPTS` property to set other Java options as well.
As described in previous sections, you must synchronize the updated PXF configuration to the Greenplum Database cluster and restart the PXF server on each segment host.
## <a id="pxf-fragcache"></a>PXF Fragment Metadata Caching
A PXF connector *Fragmenter* uses metadata from the external data source to split data into a list of fragments (blocks, files, etc.) that can be read in parallel. PXF caches the fragment metadata on a per-query basis: the first thread to access a fragment's metadata stores the information in a cache, and other threads reuse this cached metadata. Caching of this nature reduces query memory requirements for external data sources with a large number of fragments.
PXF fragment metadata caching is enabled by default. To turn off fragment metadata caching, or to re-enable it after turning it off, perform the following procedure:
1. Log in to your Greenplum Database master node:
``` shell
$ ssh gpadmin@<gpmaster>
```
2. Recall the location of the PXF user configuration directory (`$PXF_CONF`). Edit the `$PXF_CONF/conf/pxf-env.sh` file. For example:
``` shell
gpadmin@gpmaster$ vi $PXF_CONF/conf/pxf-env.sh
```
3. Locate the `PXF_FRAGMENTER_CACHE` setting in the `pxf-env.sh` file. If the setting is commented out, uncomment it, and then update the value. For example, to turn off fragment metadata caching, set the value to `false`:
``` shell
export PXF_FRAGMENTER_CACHE=false
```
3. Save the file and exit the editor.
4. Use the `pxf cluster sync` command to copy the updated `pxf-env.sh` file to the Greenplum Database cluster. For example:
``` shell
gpadmin@gpmaster$ $GPHOME/pxf/bin/pxf cluster sync
```
5. Restart PXF on each Greenplum Database segment host as described in [Restarting PXF](cfginitstart_pxf.html#restart_pxf).
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册