[
https://issues.apache.org/jira/browse/LOG4J2-431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14138418#comment-14138418
]
Remko Popma commented on LOG4J2-431:
------------------------------------
Update: user manual documentation is complete except for one outstanding change
to rewrite the initial sentence and remove the "Beta" label. Further feedback
welcome.
Next item would be a performance test report comparing this appender to
RandomAccessFile and File appender, but that is not a showstopper and may be
included in a subsequent release.
Java is supposed to be platform-independent so perhaps I worry too much, but
since memory-mapped files may have some platform-specific idiosyncrasies, I
tested on a number of platforms.
JUnit tests pass on
* 32-bit Windows XP (32 bit Oracle JVM 1.7.0_55)
* 64-bit Windows 7 (64 bit Oracle JVM 1.8.0_05 - and a few others, I'll update
the exact JVM versions later)
* 64-bit Solaris 10 (64 bit Oracle JVM 1.7.0_06-b24)
* 64-bit RHEL 5.5 (Linux 2.6.18-194.el5) with 64 bit Oracle JDK1.7.0_05-b06
* 64-bit RHEL 6.5 (Linux 2.6.32-431.el6.x86_64) with 64 bit Oracle
JDK1.7.0_05-b06 and 64 bit OpenJDK1.7.0_45 (rhel-2.4.3.3.el6-x86_64 u45-b15)
Test scenarios:
* Create new file with default region length, log less than file size. Ensure
the log file is shrunk to actually used size, and all log statements are
present in the file.
* Append to an existing file. Extend by default region length, log less than
file length. Ensure that previous data is not overwritten, new data is
appended, and the log file is shrunk to the actually used size when released.
* Create a new file with very small region length, repeatedly log to exceed
region length. Ensure all data is correctly logged and the log file is shrunk
to the actually used size when released.
More test scenarios welcome.
If no objections I will merge this into master tonight (8-12 hours from now).
> Create MemoryMappedFileAppender
> -------------------------------
>
> Key: LOG4J2-431
> URL: https://issues.apache.org/jira/browse/LOG4J2-431
> Project: Log4j 2
> Issue Type: New Feature
> Components: Appenders
> Reporter: Remko Popma
> Assignee: Remko Popma
> Priority: Minor
> Fix For: 2.1
>
> Attachments: MemoryMappedFileAppender.java,
> MemoryMappedFileAppenderTest.java, MemoryMappedFileAppenderTest.xml,
> MemoryMappedFileManager.java, MemoryMappedFileManagerTest.java
>
>
> A memory-mapped file appender may have better performance than the ByteBuffer
> + RandomAccessFile combination used by the RandomAccessFileAppender.
> *Drawbacks*
> * The drawback is that the file needs to be pre-allocated and only up to the
> file size can be mapped into memory. When the end of the file is reached the
> appender would need to extend the file and re-map.
> * Remapping is expensive (I think single-digit millisecond-range, need to
> check). For low-latency apps this kind of spike may be unacceptable so
> careful tuning is required.
> * Memory usage: If re-mapping happens too often you lose the performance
> benefits, so the memory-mapped buffer needs to be fairly large, which uses up
> memory.
> * At roll-over and shutdown the file should be truncated to immediately after
> the last written data (otherwise the user is left with a log file that ends
> in garbage).
> *Advantages*
> Measuring on a Solaris box, the difference between flushing to disk (with
> {{RandomAccessFile.write(bytes[])}}) and putting data in a MappedByteBuffer
> is about 20x: around 600ns for a ByteBuffer put and around 12-15 microseconds
> for a RandomAccessFile.write.
> (Of course different hardware and OS may give different results...)
> *Use cases*
> The difference may be most visible if {{immediateFlush}} is set to {{true}},
> which is only recommended if async loggers/appenders are not used. If
> {{immediateFlush=false}}, the large buffer used by RandomAccessFileAppender
> means you won't need to touch disk very often.
> So a MemoryMappedFileAppender is most useful in _synchronous_ logging
> scenarios, where you get the speed of writing to memory but the data is
> available on disk almost immediately. (MMap writes directly to the OS disk
> buffer.)
> In case of a application crash, the OS ensures that all data in the buffer
> will be written to disk. In case of an OS crash the data that was most
> recently added to the buffer may not be written to disk.
> Because by nature this appender would occupy a fair amount of memory, it is
> most suitable for applications running on server-class hardware with lots of
> memory available.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
