Google Cloud Pub/Sub triggers high latency on low messages throughput

Question

i'm running project which publishes messages to a PubSub topic and triggers background cloud function.

I read that with high volumes of messages, it performs well, but for lesser amounts like hundreds or even tens of messages per second, Pub/Sub may yield high latencies..

Code example to publish single message:

 const {PubSub} = require('@google-cloud/pubsub');

 const pubSubClient = new PubSub();

 async function publishMessage() {
    const topicName = 'my-topic';
    const dataBuffer = Buffer.from(data);

    const messageId = await pubSubClient.topic(topicName).publish(dataBuffer);
    console.log(`Message ${messageId} published.`);
 }

 publishMessage().catch(console.error);

Code example function triggered by PubSub:

exports.subscribe = async (message) => {
  const name = message.data
    ? Buffer.from(message.data, 'base64').toString()
    : 'World';

  console.log(`Hello, ${name}!`);
}

Cloud Function Environment Details:
Node: 8
google-cloud/pubsub: 1.6.0

The problem is when using PubSub with low throughput of messages (for example, 1 request per second) it struggles sometimes and shows incredibly high latency (up to 7-9s or more).

Is there a way or workaround to make PubSub perform well each time ( 50ms or less delay) even with small amount of incoming messages?

Answer 1

If you are always publishing to the same topic, you'd be better off keeping the object returned from pubSubClient.topic(topicName) and reusing it, whether you have a small number or large number of messages. If you want to minimize latency, you'll also want to set the maxMilliseconds property of the batching setting. By default, this is 10ms. As you have the code now, it means that every publish waits 10ms to send a message in hopes of filling the batch. Given that you create a new publisher via the topic call on every publish, you are guaranteed to always wait at least 10ms. You can set it when you call topic :

const publisher = pubSubClient.topic(topicName, {
      batching: {
        maxMessages: 100,
        maxMilliseconds: 1,
      },
    });

If after reusing the the object returned from pubSubClient.topic(topicName) and changing maxMilliseconds you are still experiencing such a delay, then you should reach out to Google Cloud Support so they can look at the specific project and topic you are using as that kind of latency is definitely not expected.