[HN Gopher] The Basics of Apache Kafka
___________________________________________________________________
The Basics of Apache Kafka
Author : the2ndfloorguy
Score : 54 points
Date : 2021-09-07 08:05 UTC (14 hours ago)
(HTM) web link (www.pankajtanwar.in)
(TXT) w3m dump (www.pankajtanwar.in)
| Railsify wrote:
| Kafka 2.8.0 removes the mandatory zookeeper dependency.
| rad_gruchalski wrote:
| This should be labelled as Show HN. One comment - towards the end
| you say
|
| > You might think that Kafka would be using queue data structure
| internally. It's not true. Kafka uses a "log" data structure. It
| is a persistent data structure which allows only appends, no
| editing, no deletion. In detail, we will cover this some other
| day.
|
| Almost immediately followed by:
|
| > Ok, I have a confession . I lied. A Kafka topic is not just a
| single queue. It's a combination of queues which helps kafka
| scale. Every queue is called partition.
| Tomte wrote:
| > This should be labelled as Show HN
|
| No, it's a simple blog post.
| rad_gruchalski wrote:
| Yes, posted by the author of the blog post.
| thunderbong wrote:
| From the FAQ [0] -
|
| > Show HN is for sharing your personal work and has special
| rules.
|
| The first line of the rules [1] -
|
| > Show HN is for something you've made that other people
| can play with. HN users can try it out, give you feedback,
| and ask questions in the thread.
|
| [0]: https://news.ycombinator.com/newsfaq.html
|
| [1]: https://news.ycombinator.com/showhn.html
| rad_gruchalski wrote:
| You're correct. I have learned something new:
|
| > Off topic: blog posts, sign-up pages, newsletters,
| lists, and other reading material. Those can't be tried
| out, so can't be Show HNs. Make a regular submission
| instead.
| moedersmooiste wrote:
| We use Kafka a lot within the company I work for. I think it's
| great. The only thing I miss is fast lookups based on some key
| and/or the ability for subscribers to only receive messages for
| certain keys.
| rad_gruchalski wrote:
| That's most likely the trade off between the queue and a log.
| The property of the log is having to look at every message.
|
| The prefixes can be somewhat done with a custom partitioner,
| the lookup requires another technology. The only way to know
| the final value under the key is to look at the complete log.
| dmead wrote:
| oddly enough, zeromq has the prefix/key style subscription.
|
| you might look at ksql for filtering purposes or just tried to
| work such things into your topics/partitions?
| brudgers wrote:
| At the computer science level, there's _What Every Software
| Engineer Should Know About Real-Time Data Unifying_ from 2013:
|
| https://engineering.linkedin.com/distributed-systems/log-wha...
|
| Reading it was an aha. At its core, the log is the simplest thing
| that might work in a lot of cases.
___________________________________________________________________
(page generated 2021-09-07 23:02 UTC)