interfaceRxErrors: "Receive errors in " + interval,
		interfaceTxBytes:  "Bytes transmitted in " + interval,
		interfaceTxErrors: "Transmit errors in " + interval,
		total:             "Total memory on the node",
		memUsed:           "Used memory on the node",
		memUsedPerc:       "Used memory percentage on the node",
		memFree:           "Free memory on the node",
		memShared:         "Shared memory on the node",

0.002 responseHeadersStart
0.082 responseHeadersEnd
0.082 responseBodyStart
0.082 responseBodyEnd
0.083 connectionReleased
0.083 callEnd
```

Notice how no connect events are fired for the second call. It reused the connection from the first request for dramatically better performance.

### EventListener.Factory

  override fun isOutputShutdown(): Boolean = delegate!!.isOutputShutdown

  @Throws(SocketException::class)
  override fun setReuseAddress(reuse: Boolean) {
    delegate!!.reuseAddress = reuse
  }

  @Throws(SocketException::class)
  override fun getReuseAddress(): Boolean = delegate!!.reuseAddress

  @Throws(SocketException::class)

   *         .putEdgeValue(BERLIN, BRUSSELS, kilometers(764))
   *         .addNode(REYKJAVIK)
   *         .build();
   * }
   *
   * <p>Builder instances can be reused; it is safe to call {@link #build} multiple times to build
   * multiple graphs in series. Each new graph contains all the elements of the ones created before
   * it.
   *
   * @since 28.0
   */

    /**
     * Sets whether the build should be resumed from the data in the resume.properties file.
     * @param resume Whether or not to resume a previous build.
     * @return This request, never {@code null}.
     */
    MavenExecutionRequest setResume(boolean resume);

    /**
     * @return Whether the build should be resumed from the data in the resume.properties file.
     */
    boolean isResume();

   * want a different order, consider using {@link #orderEntriesByValue(Comparator)}, which changes
   * this builder to sort entries by value.
   *
   * <p>Builder instances can be reused - it is safe to call {@link #buildOrThrow} multiple times to
   * build multiple bimaps in series. Each bimap is a superset of the bimaps created before it.
   *
   * @since 2.0
   */

              <placement>X</placement>
            </tag>
            <tag>
              <name>jls</name>
              <placement>X</placement>
            </tag>
            <tag>
              <name>revised</name>
              <placement>X</placement>
            </tag>
            <tag>
              <name>spec</name>
              <placement>X</placement>
            </tag>
          </tags>

    taskRunner.withLock {
      val sleepUntil = nanoTime + durationNanos
      yieldUntil { nanoTime >= sleepUntil }
    }
  }

  /**
   * Artificially stall until manually resumed by the test thread with [runTasks]. Use this to
   * simulate races in tasks that doesn't have a deterministic sequence.
   */
  fun yield() {
    taskRunner.assertLockNotHeld()
    taskRunner.withLock {

And it will save the returned value in a <abbr title="A utility/system to store computed/generated values, to reuse them instead of computing them again.">"cache"</abbr> and pass it to all the "dependants" that need it in that specific request, instead of calling the dependency multiple times for the same request.

- Key management is now centralized. There is one way to change or rotate encryption keys.
   There used to be two different mechanisms - one for regular S3 objects and one for IAM data.
- Reduced server startup time. For IAM encryption with the root credentials, MinIO had
   to use a memory-hard function (Argon2) that (on purpose) consumes a lot of memory and CPU.