/**
 * A sorted set multimap which forwards all its method calls to another sorted set multimap.
 * Subclasses should override one or more methods to modify the behavior of the backing multimap as
 * desired per the <a href="http://en.wikipedia.org/wiki/Decorator_pattern">decorator pattern</a>.
 *
 * <p><b>{@code default} method warning:</b> This class does <i>not</i> forward calls to {@code

/**
 * A {@link ListenableFuture} which forwards all its method calls to another future. Subclasses
 * should override one or more methods to modify the behavior of the backing future as desired per
 * the <a href="http://en.wikipedia.org/wiki/Decorator_pattern">decorator pattern</a>.
 *
 * <p>Most subclasses can just use {@link SimpleForwardingListenableFuture}.
 *
 * @author Shardul Deo
 * @since 4.0
 */

# unmarkComment: >
#   Your comment here.

# Comment to post when closing a stale Issue or Pull Request.
# closeComment: >
#   Your comment here.

# Limit the number of actions per hour, from 1-30. Default is 30
limitPerRun: 1

# Limit to only `issues` or `pulls`
# only: issues

# Optionally, specify configuration settings that are specific to just 'issues' or 'pulls':
# pulls:

*.threaddump
# ps output for cleaning up leaking Java processes
*.psoutput

# oh-my-zsh gradle plugin
.gradletasknamecache

# Added GE support maven support to the maven build in .teamcity, per the GE docs, this dir is NOT to be committed
.teamcity/.mvn/.develocity/
/discoclient.properties

# Ignore local configuration files for asdf, allowing the JDK to be configured for project (https://asdf-vm.com)

	}
	const payloadSize = 512
	rng := rand.New(rand.NewSource(time.Now().UnixNano()))
	payload := make([]byte, payloadSize)
	_, err = rng.Read(payload)
	errFatal(err)

	// Wait for all to connect
	// Parallel writes per server.
	b.Run("bytes", func(b *testing.B) {
		for par := 1; par <= 32; par *= 2 {
			b.Run("par="+strconv.Itoa(par*runtime.GOMAXPROCS(0)), func(b *testing.B) {
				defer timeout(60 * time.Second)()

Let's imagine that you have some **machine learning models** that you want to use to handle requests. 🤖

The same models are shared among requests, so, it's not one model per request, or one per user or something similar.

Let's imagine that loading the model can **take quite some time**, because it has to read a lot of **data from disk**. So you don't want to do it for every request.

/**
 * A list iterator which forwards all its method calls to another list iterator. Subclasses should
 * override one or more methods to modify the behavior of the backing iterator as desired per the <a
 * href="http://en.wikipedia.org/wiki/Decorator_pattern">decorator pattern</a>.
 *
 * <p><b>{@code default} method warning:</b> This class forwards calls to <i>only some</i> {@code

import javax.annotation.CheckForNull;

/**
 * A concurrent map which forwards all its method calls to another concurrent map. Subclasses should
 * override one or more methods to modify the behavior of the backing map as desired per the <a
 * href="http://en.wikipedia.org/wiki/Decorator_pattern">decorator pattern</a>.
 *
 * <p><b>{@code default} method warning:</b> This class forwards calls to <i>only some</i> {@code

		memCache:          "Cache memory on the node",
		memAvailable:      "Available memory on the node",
		readsPerSec:       "Reads per second on a drive",
		writesPerSec:      "Writes per second on a drive",
		readsKBPerSec:     "Kilobytes read per second on a drive",
		writesKBPerSec:    "Kilobytes written per second on a drive",
		readsAwait:        "Average time for read requests to be served on a drive",

/**
 * A list iterator which forwards all its method calls to another list iterator. Subclasses should
 * override one or more methods to modify the behavior of the backing iterator as desired per the <a
 * href="http://en.wikipedia.org/wiki/Decorator_pattern">decorator pattern</a>.
 *
 * <p><b>{@code default} method warning:</b> This class forwards calls to <i>only some</i> {@code