FastDateFormat is not very fast

[wsargent]

I've been doing some profiling using the chronicle appender with JSON output. It looks like the FastDateFormat used by FormattedTimestampJsonProvider is creating a calendar with `newCalendar and formatting it every time, which creates unnecessary garbage and is a good chunk of the JSON formatting.

https://github.com/apache/commons-lang/blob/master/src/main/java/org/apache/commons/lang3/time/FastDatePrinter.java#L444

It might be good to use DateTimeFormatter.ISO_DATETIME_FORMAT directly and compare it.

logback-appender-CHRONICLE-2  Runnable CPU usage on sample: 236ms
  net.logstash.logback.encoder.org.apache.commons.lang3.time.FastDatePrinter.format(long) FastDatePrinter.java:446
  net.logstash.logback.encoder.org.apache.commons.lang3.time.FastDateFormat.format(long) FastDateFormat.java:428
  net.logstash.logback.composite.FormattedTimestampJsonProvider$PatternTimestampWriter.getTimestampAsString(long) FormattedTimestampJsonProvider.java:101
  net.logstash.logback.composite.FormattedTimestampJsonProvider$PatternTimestampWriter.writeTo(JsonGenerator, String, long) FormattedTimestampJsonProvider.java:96
  net.logstash.logback.composite.FormattedTimestampJsonProvider.writeTo(JsonGenerator, DeferredProcessingAware) FormattedTimestampJsonProvider.java:149
  net.logstash.logback.composite.JsonProviders.writeTo(JsonGenerator, DeferredProcessingAware) JsonProviders.java:77
  net.logstash.logback.composite.CompositeJsonFormatter.writeEventToGenerator(JsonGenerator, DeferredProcessingAware) CompositeJsonFormatter.java:193
  net.logstash.logback.composite.CompositeJsonFormatter.writeEventToOutputStream(DeferredProcessingAware, OutputStream) CompositeJsonFormatter.java:170
  net.logstash.logback.encoder.CompositeJsonEncoder.encode(DeferredProcessingAware) CompositeJsonEncoder.java:122
  net.logstash.logback.encoder.CompositeJsonEncoder.encode(Object) CompositeJsonEncoder.java:34
  net.openhft.chronicle.logger.logback.UnsynchronizedChronicleAppender.append(ILoggingEvent) UnsynchronizedChronicleAppender.java:34
  net.openhft.chronicle.logger.logback.UnsynchronizedChronicleAppender.append(Object) UnsynchronizedChronicleAppender.java:30
  ch.qos.logback.core.UnsynchronizedAppenderBase.doAppend(Object) UnsynchronizedAppenderBase.java:84
  ch.qos.logback.core.spi.AppenderAttachableImpl.appendLoopOnAppenders(Object) AppenderAttachableImpl.java:51
  net.logstash.logback.appender.DelegatingAsyncDisruptorAppender$DelegatingEventHandler.onEvent(AsyncDisruptorAppender$LogEvent, long, boolean) DelegatingAsyncDisruptorAppender.java:55
  net.logstash.logback.appender.DelegatingAsyncDisruptorAppender$DelegatingEventHandler.onEvent(Object, long, boolean) DelegatingAsyncDisruptorAppender.java:51
  net.logstash.logback.appender.AsyncDisruptorAppender$EventClearingEventHandler.onEvent(AsyncDisruptorAppender$LogEvent, long, boolean) AsyncDisruptorAppender.java:339
  net.logstash.logback.appender.AsyncDisruptorAppender$EventClearingEventHandler.onEvent(Object, long, boolean) AsyncDisruptorAppender.java:327
  net.logstash.logback.encoder.com.lmax.disruptor.BatchEventProcessor.processEvents() BatchEventProcessor.java:168
  net.logstash.logback.encoder.com.lmax.disruptor.BatchEventProcessor.run() BatchEventProcessor.java:125
  java.util.concurrent.Executors$RunnableAdapter.call() Executors.java:515
  java.util.concurrent.FutureTask.run() FutureTask.java:264
  java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run() ScheduledThreadPoolExecutor.java:304
  java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor$Worker) ThreadPoolExecutor.java:1128
  java.util.concurrent.ThreadPoolExecutor$Worker.run() ThreadPoolExecutor.java:628
  java.lang.Thread.run() Thread.java:834

[philsttr]

Hey @wsargent ,

Luckily the switch to DateTimeFormatter is already implemented in #377, but not merged yet. It's not fully backwards compatible, so it will require a major version bump. I was holding off on merging it to see if there were any other major (non-backwards compatible) changes that needed to be made, that could be grouped with it. And I wasn't in a rush to merge it, because FastDateFormat has worked for a long time. If it is causing serious problems, I'll consider merging it and bumping the major version soon.

[wsargent]

@philsttr I think there's a smaller workaround, which is to use a format(calendar, buf) and reuse the calendar instances. Not sure how much of an improvement that would be as I haven't benchmarked it.

https://github.com/apache/commons-lang/blob/master/src/main/java/org/apache/commons/lang3/time/FastDatePrinter.java#L537