Streaming Audio: Apache Kafka® & Real-Time Data

Improving Apache Kafka Scalability and Elasticity with Tiered Storage


Listen Later

What happens when you need to store more than a few petabytes of data? Rittika Adhikari (Software Engineer, Confluent) discusses how her team implemented tiered storage, a method for improving the scalability and elasticity of data storage in Apache Kafka®. She also explores the motivating factors for building it in the first place: cost, performance, and manageability. 

Before Tiered Storage, there was no real way to retain Kafka data indefinitely. Because of the tight coupling between compute and storage, users were forced to use different tools to access cold and hot data. Additionally, the cost of re-replication was prohibitive because Kafka had to process large amounts of data rather than small hot sets.

As a member of the Kafka Storage Foundations team, Rittika explains to Kris Jenkins how her team initially considered a Kafka data lake but settled on a more cost-effective method – tiered storage. With tiered storage, one tier handles elasticity and throughput for long-term storage, while the other tier is dedicated to high-cost, low-latency, short-term storage. Before, re-replication impacted all brokers, slowing down performance because it required more replication cycles. By decoupling compute and storage, they now only replicate the hot set rather than weeks of data. 

Ultimately, this tiered storage method broke down the barrier between compute and storage by separating data into multiple tiers across the cloud. This allowed for better scalability and elasticity that reduced operational toil. 

In preparation for a broader rollout to customers who heavily rely on compacted topics, Rittika’s team will be implementing tier compaction to support tiering of compacted topics. The goal is to have the partition leader perform compaction. This will substantially reduce compaction costs (CPU/disk) because the number of replicas compacting is significantly smaller. It also protects the broker resource consumption through a new compaction algorithm and throttling. 

EPISODE LINKS

  • Jun Rao explains: What is Tiered Storage?
  • Enabling Tiered Storage
  • Infinite Storage in Confluent Platform
  • Kafka Storage and Processing Fundamentals
  • KIP-405: Kafka Tiered Storage
  • Optimizing Apache Kafka’s Internals with Its Co-Creator Jun Rao
  • Watch the video version of this podcast
  • Kris Jenkins’ Twitter
  • Streaming Audio Playlist 
  • Join the Confluent Community
  • Learn more with Kafka tutorials, resources, and guides at Confluent Developer
  • Live demo: Intro to Event-Driven Microservices with Confluent
  • Use PODCAST100 to get an additional $100 of free Confluent Cloud usage (details)  
...more
View all episodesView all episodes
Download on the App Store

Streaming Audio: Apache Kafka® & Real-Time DataBy Confluent, founded by the original creators of Apache Kafka®

  • 4.8
  • 4.8
  • 4.8
  • 4.8
  • 4.8

4.8

44 ratings


More shows like Streaming Audio: Apache Kafka® & Real-Time Data

View all
Planet Money by NPR

Planet Money

30,892 Listeners

The Changelog: Software Development, Open Source by Changelog Media

The Changelog: Software Development, Open Source

285 Listeners

Talk Python To Me by Michael Kennedy

Talk Python To Me

586 Listeners

Software Engineering Daily by Software Engineering Daily

Software Engineering Daily

631 Listeners

AWS Podcast by Amazon Web Services

AWS Podcast

201 Listeners

DataFramed by DataCamp

DataFramed

268 Listeners

Tech Lead Journal by Henry Suryawirawan

Tech Lead Journal

12 Listeners

System Design by Wes and Kevin

System Design

93 Listeners

Postgres FM by Nikolay Samokhvalov and Michael Christofides

Postgres FM

20 Listeners

Kubernetes for Humans by Komodor

Kubernetes for Humans

2 Listeners

Learn System Design by Ben Kitchell

Learn System Design

32 Listeners