Files
spring-cloud-static/spring-cloud-stream-binder-kafka/3.0.4.RELEASE/reference/html/partitions.html
2020-04-22 16:43:12 +00:00

241 lines
7.7 KiB
HTML

<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<!--[if IE]><meta http-equiv="X-UA-Compatible" content="IE=edge"><![endif]-->
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta name="generator" content="Asciidoctor 1.5.8">
<title>Partitioning with the Kafka Binder</title>
<link rel="stylesheet" href="css/spring.css">
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.min.css">
<style>
.hidden {
display: none;
}
.switch {
border-width: 1px 1px 0 1px;
border-style: solid;
border-color: #7a2518;
display: inline-block;
}
.switch--item {
padding: 10px;
background-color: #ffffff;
color: #7a2518;
display: inline-block;
cursor: pointer;
}
.switch--item:not(:first-child) {
border-width: 0 0 0 1px;
border-style: solid;
border-color: #7a2518;
}
.switch--item.selected {
background-color: #7a2519;
color: #ffffff;
}
</style>
<script src="https://cdnjs.cloudflare.com/ajax/libs/zepto/1.2.0/zepto.min.js"></script>
<script type="text/javascript">
function addBlockSwitches() {
$('.primary').each(function() {
primary = $(this);
createSwitchItem(primary, createBlockSwitch(primary)).item.addClass("selected");
primary.children('.title').remove();
});
$('.secondary').each(function(idx, node) {
secondary = $(node);
primary = findPrimary(secondary);
switchItem = createSwitchItem(secondary, primary.children('.switch'));
switchItem.content.addClass('hidden');
findPrimary(secondary).append(switchItem.content);
secondary.remove();
});
}
function createBlockSwitch(primary) {
blockSwitch = $('<div class="switch"></div>');
primary.prepend(blockSwitch);
return blockSwitch;
}
function findPrimary(secondary) {
candidate = secondary.prev();
while (!candidate.is('.primary')) {
candidate = candidate.prev();
}
return candidate;
}
function createSwitchItem(block, blockSwitch) {
blockName = block.children('.title').text();
content = block.children('.content').first().append(block.next('.colist'));
item = $('<div class="switch--item">' + blockName + '</div>');
item.on('click', '', content, function(e) {
$(this).addClass('selected');
$(this).siblings().removeClass('selected');
e.data.siblings('.content').addClass('hidden');
e.data.removeClass('hidden');
});
blockSwitch.append(item);
return {'item': item, 'content': content};
}
$(addBlockSwitches);
</script>
</head>
<body class="book toc2 toc-left">
<div id="header">
<div id="toc" class="toc2">
<div id="toctitle">Table of Contents</div>
<ul class="sectlevel2">
<li><a href="#_partitioning_with_the_kafka_binder">Partitioning with the Kafka Binder</a></li>
</ul>
</div>
</div>
<div id="content">
<div class="sect2">
<h3 id="_partitioning_with_the_kafka_binder"><a class="link" href="#_partitioning_with_the_kafka_binder">Partitioning with the Kafka Binder</a></h3>
<div class="paragraph">
<p>Apache Kafka supports topic partitioning natively.</p>
</div>
<div class="paragraph">
<p>Sometimes it is advantageous to send data to specific partitions&#8201;&#8212;&#8201;for example, when you want to strictly order message processing (all messages for a particular customer should go to the same partition).</p>
</div>
<div class="paragraph">
<p>The following example shows how to configure the producer and consumer side:</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="highlightjs highlight"><code class="language-java hljs" data-lang="java">@SpringBootApplication
@EnableBinding(Source.class)
public class KafkaPartitionProducerApplication {
private static final Random RANDOM = new Random(System.currentTimeMillis());
private static final String[] data = new String[] {
"foo1", "bar1", "qux1",
"foo2", "bar2", "qux2",
"foo3", "bar3", "qux3",
"foo4", "bar4", "qux4",
};
public static void main(String[] args) {
new SpringApplicationBuilder(KafkaPartitionProducerApplication.class)
.web(false)
.run(args);
}
@InboundChannelAdapter(channel = Source.OUTPUT, poller = @Poller(fixedRate = "5000"))
public Message&lt;?&gt; generate() {
String value = data[RANDOM.nextInt(data.length)];
System.out.println("Sending: " + value);
return MessageBuilder.withPayload(value)
.setHeader("partitionKey", value)
.build();
}
}</code></pre>
</div>
</div>
<div class="listingblock">
<div class="title">application.yml</div>
<div class="content">
<pre class="highlightjs highlight"><code class="language-yaml hljs" data-lang="yaml">spring:
cloud:
stream:
bindings:
output:
destination: partitioned.topic
producer:
partition-key-expression: headers['partitionKey']
partition-count: 12</code></pre>
</div>
</div>
<div class="admonitionblock important">
<table>
<tr>
<td class="icon">
<i class="fa icon-important" title="Important"></i>
</td>
<td class="content">
The topic must be provisioned to have enough partitions to achieve the desired concurrency for all consumer groups.
The above configuration supports up to 12 consumer instances (6 if their <code>concurrency</code> is 2, 4 if their concurrency is 3, and so on).
It is generally best to &#8220;over-provision&#8221; the partitions to allow for future increases in consumers or concurrency.
</td>
</tr>
</table>
</div>
<div class="admonitionblock note">
<table>
<tr>
<td class="icon">
<i class="fa icon-note" title="Note"></i>
</td>
<td class="content">
The preceding configuration uses the default partitioning (<code>key.hashCode() % partitionCount</code>).
This may or may not provide a suitably balanced algorithm, depending on the key values.
You can override this default by using the <code>partitionSelectorExpression</code> or <code>partitionSelectorClass</code> properties.
</td>
</tr>
</table>
</div>
<div class="paragraph">
<p>Since partitions are natively handled by Kafka, no special configuration is needed on the consumer side.
Kafka allocates partitions across the instances.</p>
</div>
<div class="paragraph">
<p>The following Spring Boot application listens to a Kafka stream and prints (to the console) the partition ID to which each message goes:</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="highlightjs highlight"><code class="language-java hljs" data-lang="java">@SpringBootApplication
@EnableBinding(Sink.class)
public class KafkaPartitionConsumerApplication {
public static void main(String[] args) {
new SpringApplicationBuilder(KafkaPartitionConsumerApplication.class)
.web(false)
.run(args);
}
@StreamListener(Sink.INPUT)
public void listen(@Payload String in, @Header(KafkaHeaders.RECEIVED_PARTITION_ID) int partition) {
System.out.println(in + " received from partition " + partition);
}
}</code></pre>
</div>
</div>
<div class="listingblock">
<div class="title">application.yml</div>
<div class="content">
<pre class="highlightjs highlight"><code class="language-yaml hljs" data-lang="yaml">spring:
cloud:
stream:
bindings:
input:
destination: partitioned.topic
group: myGroup</code></pre>
</div>
</div>
<div class="paragraph">
<p>You can add instances as needed.
Kafka rebalances the partition allocations.
If the instance count (or <code>instance count * concurrency</code>) exceeds the number of partitions, some consumers are idle.</p>
</div>
</div>
</div>
<script type="text/javascript" src="js/tocbot/tocbot.min.js"></script>
<script type="text/javascript" src="js/toc.js"></script>
<link rel="stylesheet" href="js/highlight/styles/atom-one-dark-reasonable.min.css">
<script src="js/highlight/highlight.min.js"></script>
<script>hljs.initHighlighting()</script>
</body>
</html>