Amazon SQS vs Kafka: What are the differences?
Developers describe Amazon SQS as "Fully managed message queuing service". Transmit any volume of data, at any level of throughput, without losing messages or requiring other services to be always available. With SQS, you can offload the administrative burden of operating and scaling a highly available messaging cluster, while paying a low price for only what you use. On the other hand, Kafka is detailed as "Distributed, fault tolerant, high throughput pub-sub messaging system". Kafka is a distributed, partitioned, replicated commit log service. It provides the functionality of a messaging system, but with a unique design.
Amazon SQS and Kafka can be categorized as "Message Queue" tools.
Some of the features offered by Amazon SQS are:
- A queue can be created in any region.
- The message payload can contain up to 256KB of text in any format. Each 64KB ‘chunk’ of payload is billed as 1 request. For example, a single API call with a 256KB payload will be billed as four requests.
- Messages can be sent, received or deleted in batches of up to 10 messages or 256KB. Batches cost the same amount as single messages, meaning SQS can be even more cost effective for customers that use batching.
On the other hand, Kafka provides the following key features:
- Written at LinkedIn in Scala
- Used by LinkedIn to offload processing of all page and other views
- Defaults to using persistence, uses OS disk cache for hot data (has higher throughput then any of the above having persistence enabled)
"Easy to use, reliable" is the top reason why over 45 developers like Amazon SQS, while over 95 developers mention "High-throughput" as the leading cause for choosing Kafka.
Kafka is an open source tool with 12.7K GitHub stars and 6.81K GitHub forks. Here's a link to Kafka's open source repository on GitHub.
Uber Technologies, Spotify, and Slack are some of the popular companies that use Kafka, whereas Amazon SQS is used by Medium, Lyft, and Coursera. Kafka has a broader approval, being mentioned in 509 company stacks & 470 developers stacks; compared to Amazon SQS, which is listed in 384 company stacks and 103 developer stacks.
What is Amazon SQS?
What is Kafka?
Need advice about which tool to choose?Ask the StackShare community!
Sign up to add, upvote and see more prosMake informed product decisions
Sign up to get full access to all the companiesMake informed product decisions
Sign up to get full access to all the tool integrationsMake informed product decisions
Why we built Marmaray, an open source generic data ingestion and dispersal framework and library for Apache Hadoop :
Built and designed by our Hadoop Platform team, Marmaray is a plug-in-based framework built on top of the Hadoop ecosystem. Users can add support to ingest data from any source and disperse to any sink leveraging the use of Apache Spark . The name, Marmaray, comes from a tunnel in Turkey connecting Europe and Asia. Similarly, we envisioned Marmaray within Uber as a pipeline connecting data from any source to any sink depending on customer preference:
(Direct GitHub repo: https://github.com/uber/marmaray Kafka Kafka Manager )
I use Kafka because it has almost infinite scaleability in terms of processing events (could be scaled to process hundreds of thousands of events), great monitoring (all sorts of metrics are exposed via JMX).
Downsides of using Kafka are: - you have to deal with Zookeeper - you have to implement advanced routing yourself (compared to RabbitMQ it has no advanced routing)
The question for which Message Queue to use mentioned "availability, distributed, scalability, and monitoring". I don't think that this excludes many options already. I does not sound like you would take advantage of Kafka's strengths (replayability, based on an even sourcing architecture). You could pick one of the AMQP options.
I would recommend the RabbitMQ message broker, which not only implements the AMQP standard 0.9.1 (it can support 1.x or other protocols as well) but has also several very useful extensions built in. It ticks the boxes you mentioned and on top you will get a very flexible system, that allows you to build the architecture, pick the options and trade-offs that suite your case best.
For more information about RabbitMQ, please have a look at the linked markdown I assembled. The second half explains many configuration options. It also contains links to managed hosting and to libraries (though it is missing Python's - which should be Puka, I assume).
I used Kafka originally because it was mandated as part of the top-level IT requirements at a Fortune 500 client. What I found was that it was orders of magnitude more complex ...and powerful than my daily Beanstalkd , and far more flexible, resilient, and manageable than RabbitMQ.
So for any case where utmost flexibility and resilience are part of the deal, I would use Kafka again. But due to the complexities involved, for any time where this level of scalability is not required, I would probably just use Beanstalkd for its simplicity.
I tend to find RabbitMQ to be in an uncomfortable middle place between these two extremities.
In the beginning we thought we wanted to start using something like RabbitMQ or maybe Kafka or maybe ActiveMQ. Back then we only had a few developers and no ops people. That has changed now, but we didn't really look forward to setting up a queuing cluster and making sure that all works.
What we did instead was we looked at what services Amazon offers to see if we can use those to build our own messaging system within those services. That's basically what we did. We wrote some clients in Ruby that can basically do the entire orchestration for us, and we run all our messaging on both SNS and SQS. Basically what you can do in Amazon services is you can use Amazon Simple Notification Service, so SNS, for creating topics and you can use queues to subscribe to these topics. That's basically all you need for a messaging system. You don't have to worry about scalability at all. That's what really appealed to us.
Front-end messages are logged to Kafka by our API and application servers. We have batch processing (on the middle-left) and real-time processing (on the middle-right) pipelines to process the experiment data. For batch processing, after daily raw log get to s3, we start our nightly experiment workflow to figure out experiment users groups and experiment metrics. We use our in-house workflow management system Pinball to manage the dependencies of all these MapReduce jobs.
This isn't exactly low-latency (10s to 100s of milliseconds), but it has good throughput and a simple API. There is good reliability, and there is no configuration necessary to get up and running. A hosted queue is important when trying to move fast.
SQS is the bridge between our new Lambda services and our incumbent Rails applications. Extremely easy to use when you're already using other AWS infrastructure.
Building out real-time streaming server to present data insights to Coolfront Mobile customers and internal sales and marketing teams.
Primary message queue. Enqueueing operations revert to a local file-system-based queue when SQS is unavailable.
I can't afford to lose data if Dynamo throttles my writes, so everything goes into a message queue first.