Ververica Blog
ververica.com · Company blog · on HN since 2019
Ververica, the company founded by the original creators of Apache Flink, writes about stream processing, Flink use cases and its platform.
What HN saysHN links it for Flink benchmarks and use cases; commenters suspect vendor bias in the Flink versus Databricks comparison and mention a past Databricks benchmark mistake.
16HN people shared it
4linked it in comments
21stories submitted
39story points
2 years agolast shared
2019people per year2026
Most-shared posts
- 1Curious case broken benchmark revisiting apache flink vs databricks runtime2 people · first shared 2019
- 2Yelp uses Flink for predicting store visits in real time1 people · 4 points · first shared 2019 · discussion on HN
- 3Monitoring Apache Flink Applications 1011 people · 4 points · first shared 2019 · discussion on HN
- 4High-throughput, low-latency, and exactly-once stream processing (Apache Flink)1 people · 3 points · first shared 2020 · discussion on HN
- 5Netflix Uses Flink for Scale Data Processing – Flink Forward Preview1 people · 3 points · first shared 2019 · discussion on HN
- 6Flux Capacitor, Huh? Temporal Tables and Joins in Streaming SQL1 people · 3 points · first shared 2019 · discussion on HN
- 7Apache Flink at Scale: Extending the Yahoo Streaming Benchmark1 people · 2 points · first shared 2023 · discussion on HN
- 8How to manage your RocksDB memory size in Apache Flink1 people · 2 points · first shared 2019 · discussion on HN
- 9A Farewell Message1 people · 2 points · first shared 2022 · discussion on HN
- 10OpenSSL in Ververica Platform 2.1 improves your Flink performance1 people · 2 points · first shared 2020 · discussion on HN
- 11Real-Time Performance Monitoring with Flink SQL1 people · 2 points · first shared 2022 · discussion on HN
- 12Announcing Stateful Functions: Distributed State, Uncomplicated1 people · 2 points · first shared 2019 · discussion on HN
- 13Data Artisans changes its name to Ververica1 people · 2 points · first shared 2019 · discussion on HN
- 14Announcing Ververica Platform 1.4.0 with Support for Apache Flink 1.8.01 people · 2 points · first shared 2019 · discussion on HN
- 15Smart IoT use case with open source Kafka, Flink and CrateDB1 people · 1 points · first shared 2020 · discussion on HN
What HN people wrote when they linked it
rehevkor5 · 2023-02-09
Along similar lines, I found Flink's blog posts to be good reading. Some examples:
https://flink.apache.org/news/2015/12/04/Introducing-windows.html
https://www.ververica.com/blog/session-windowing-in-flink
https://flink.apache.org/news/2020/07/30/demo-fraud-detection-3.html
glogla · 2021-11-13
What do you know, here's an article[1] from 2017 about Databricks making an unfortunate mistake that showed Spark Streaming (which they sell) as a better streaming platform to Flink (which they don't sell).
I really hope this is not the case again.
(yes, I understand my sarcasm is unneeded, I couldn't help myself)
[1]: https://www.ververica.com/blog/curious-case-broken-benchmark-revisitin…
jtagx · 2021-02-02
Ververica | Product Manager, DevOps, Junior/Senior Engineers, Technical Writer | Full Time | Berlin, Germany or EU Remote | https://www.ververica.com/careers
We are the original creators of Apache Flink, the open-source unified batch/stream processing system that powers applications in all types of companies, from tech giants like Alibaba, Amazon, or Netflix, to traditional enterprises like banks or telcos. We contribute heavily to Apache Flink while building enterprise-grade products on top of Flink.
jdm2212 · 2019-08-25
Here's a benchmark of KStreams and Flink [1]. Note that the Flink vs Spark comparison is disputed [2], but both Flink and Spark are several orders of magnitude faster than KStreams. This is inevitable given KStreams architecture -- it stores all its state in Kafka rather than in a data store and with data structures optimized for the use case and doesn't do much coordination among workers. KStreams is there if you want streaming semantics on top of a small-ish Kafka topic you own, but don't care too much about perf. Deploying and maintaining Flink is a much bigger hassle than KStreams -- you need DevOps support to get Flink running, whereas KStreams runs (albeit quite slowly) inside your application with no new state store needed.
Confluent has a good discussion of the ownership issue (DevOps for Flink, devs for KStreams) here [3] though they seriously downplay the huge gap in perf.
[1] https://databricks.com/blog/2017/10/11/benchmarking-structured-streami…
[2] https://www.ververica.com/blog/curious-case-broken-benchmark-revisitin…
[3] https://www.confluent.io/blog/apache-flink-apache-kafka-streams-compar…